FP Koenig Eagly 2011 PDF

Psychological Bulletin
2011, Vol. 137, No. 4, 616 642
2011 American Psychological Association

0033-2909/11/$12.00 DOI: 10.1037/a0023557
Are Leader Stereotypes Masculine?

A Meta-Analysis of Three Research Paradigms
Anne M. Koenig
Alice H. Eagly
University of San Diego
Northwestern University
Abigail A. Mitchell
Tiina Ristikari
Nebraska Wesleyan University
University of Tampere
This meta-analysis examined the extent to which stereotypes of leaders are culturally masculine. The
primary studies fit into 1 of 3 paradigms: (a) In Scheins (1973) think managerthink male paradigm, 40
studies with 51 effect sizes compared the similarity of male and leader stereotypes and the similarity of
female and leader stereotypes; (b) in Powell and Butterfields (1979) agency communion paradigm, 22
studies with 47 effect sizes compared stereotypes of leaders agency and communion; and (c) in Shinars
(1975) masculinityfemininity paradigm, 7 studies with 101 effect sizes represented stereotypes of
leadership-related occupations on a single masculinityfemininity dimension. Analyses implemented
appropriate random and mixed effects models. All 3 paradigms demonstrated overall masculinity of
leader stereotypes: (a) In the think managerthink male paradigm, intraclass correlation ! .25 for
the womenleaders similarity and intraclass correlation ! .62 for the menleaders similarity; (b) in the
agency communion paradigm, g ! 1.55, indicating greater agency than communion; and (c) in the
masculinityfemininity paradigm, g ! 0.92, indicating greater masculinity than the androgynous scale
midpoint. Subgroup and meta-regression analyses indicated that this masculine construal of leadership
has decreased over time and was greater for male than female research participants. In addition,
stereotypes portrayed leaders as less masculine in educational organizations than in other domains and
in moderate- than in high-status leader roles. This article considers the relation of these findings to Eagly
and Karaus (2002) role congruity theory, which proposed contextual influences on the incongruity
between stereotypes of women and leaders. The implications for prejudice against women leaders are
also considered.
Keywords: leadership, management, gender stereotypes, gender roles, meta-analysis
qualities (e.g., assertive, competitive) that they believe are required

for success as a leader (Eagly & Carli, 2007). Given that agentic
qualities are ascribed more to men than women (e.g., Gallup News
Service, 2001; Spence & Buckner, 2000), leadership is generally
associated with masculinity. This article reports a meta-analysis of
research that has assessed the cultural masculinity of leader stereotypes and explores the conditions under which this masculinity
is more or less pronounced.
The characteristics that people commonly ascribe to women,

men, and leaders contribute to the challenges that women face in
obtaining leadership roles and performing well in them.1 Cultural
stereotypes can make it seem that women do not have what it takes
for important leadership roles. This cultural mismatch, or role
incongruity, between women and the perceived demands of leadership underlies biased evaluations of women as leaders (Eagly &
Karau, 2002). Fueling this mismatch is an inconsistency between
the predominantly communal qualities (e.g., nice, compassionate)
that people associate with women and the predominantly agentic
The Importance of Stereotypes to Womens Access

to Leadership
Stereotypes often are a potent barrier to womens advancement
to positions of leadership. This assertion is the consensus view not
This article was published Online First May 30, 2011.

Anne M. Koenig, Department of Psychological Sciences, University of
San Diego; Alice H. Eagly, Department of Psychology, Northwestern
University; Abigail A. Mitchell, Department of Psychology, Nebraska
Wesleyan University; Tiina Ristikari, Department of Social Research,
University of Tampere, Tampere, Finland.
Anne M. Koenig was supported in part by the National Institutes of
Health under a Ruth L. Kirschstein National Research Service Award
(1 F31 MH074251).
Correspondence concerning this article should be addressed to Anne M.
Koenig, Department of Psychological Sciences, 5998 Alcala Park, University of San Diego, San Diego, CA 92110. E-mail: akoenig@sandiego.edu
1
In this article, we use the terms leader and manager interchangeably.
Although a distinction between leadership and management is useful in
some contexts (e.g., Bennis, 1989), the research that we review does not
allow leader and managerial roles or functions to be separated. Also, we
use the terms sex and sexes to denote the grouping of people into female
and male categories. The term gender refers to the meanings that societies
and individuals ascribe to these female and male categories. We do not
intend to use these terms to give priority to any class of causes that may
underlie sex and gender effects (see Wood & Eagly, 2010).
616
MASCULINITY OF LEADER STEREOTYPES
only of social and organizational psychologists (e.g., Glick &

Fiske, 2007; Heilman & Parks-Stamm, 2007; S. K. Johnson, Murphy, Zewdie, & Reichard, 2008), but also of women who have
substantial experience as leaders. For example, a survey of 705
women at the vice president level and above in Fortune 1,000
corporations found that 72% agreed or strongly agreed that stereotypes about womens roles and abilities are a barrier to womens advancement to the highest levels (Wellington, Kropf, &
Gerkovich, 2003).
The damaging effects of stereotypes for women as leaders do
not stem from beliefs about women that are mainly negative. On
the contrary, consistent with the women-are-wonderful effect (Eagly & Mladinic, 1994; Langford & MacKinnon, 2000), women are
regarded as the nicer, kinder sex and thus have a cultural stereotype that is in general more positive than that of men. Nevertheless, women often experience workplace discrimination (Heilman
& Eagly, 2008). Theorists have resolved this paradox by reasoning
that it is not the evaluative content of the stereotype of women but
its mismatch with many desirable work roles that underlies biased
evaluations in many employment settings (e.g., Eagly & Karau,
2002; Heilman, 2001; Hogue & Lord, 2007; Lyness & Heilman,
2006).
In Eagly and Karaus (2002) role congruity model, the mismatch
that produces biased evaluation is between stereotypes of women
(e.g., Spence & Buckner, 2000) and stereotypes of leaders (e.g.,
Epitropaki & Martin, 2004; Lord & Maher, 1993; Offermann,
Kennedy, & Wirtz, 1994). In this model, stereotype content is
framed in terms of agency and communion, with communion the
more important theme in the female stereotype and agency the
more important theme in both leader and male stereotypes. Men
are therefore seen as more similar to the leader stereotype than
women are, producing disadvantage for women. In Heilmans
(1983, 2001) broader lack-of-fit model, to the extent that a workplace role is inconsistent with the attributes ascribed to an individual, she or he would suffer from perceived lack of fit, producing
increased expectations of failure and decreased expectations of
success. The incongruity between construals of women and leaders
is thus one type of lack of fit.
These communal and agentic meanings ascribed to women and
men form a constant backdrop to social interaction, coloring the
judgments made about people encountered in organizations and
other contexts (Wood & Eagly, 2010). In both the role congruity
(Eagly & Karau, 2002) and the lack-of-fit models (Heilman, 2001),
these beliefs comprise two kinds of expectations or norms: descriptive beliefs (or stereotypes), which are consensual expectations about what members of a social group actually do, and
injunctive (or prescriptive) beliefs, which are consensual expectations about what group members ought to do or ideally would do
(Cialdini & Trost, 1998). Therefore, women are the targets of two
forms of prejudice against them as leaders: a deficit in the ascription of leadership ability to them and, compared to that of men, a
less favorable evaluation of their agentic leadership behavior. In
other words, descriptively, women seem less usual or natural in
most leadership roles; and prescriptively, women often seem inappropriate or presumptuous when they display the agentic behavior often required by these roles (see also Burgess & Borgida,
1999).
Because individuals are commonly assimilated to group stereotypes (e.g., von Hippel, Sekaquaptewa, & Vargas, 1995), an eval-
617
uative penalty is exacted from women leaders or potential leaders

regardless of whether they possess the qualities appropriate for
leadership roles. This penalty consists of unfavorable performance
expectations, which in turn enable biased judgments and less
favorable evaluations (Eagly, Makhijani, & Klonsky, 1992; Lyness
& Heilman, 2006). Discrimination is the behavioral outcome of
these processes (Eagly & Diekman, 2005).
How strong is the evidence for the masculinity of leader stereotypes? Partial reviews of research on this question have substantiated the claim about leaders cultural masculinity (e.g., Duehr &
Bono, 2006; Heilman, Block, Martell, & Simon, 1989; Powell,
Butterfield, & Parent, 2002; Schein, 2001). However, these reviews have not identified all of the relevant research paradigms or
accessed all of the available studies within each paradigm or
estimated the magnitude of the effects. Our integration of the
research literature remedies these omissions and tests the propositions of role congruity theory concerning the conditions under
which incongruity between the leader and female gender stereotype is stronger or weaker (Eagly & Karau, 2002).
Three Paradigms for Examining the Masculinity of

Leader Roles
Research in three separate paradigms has addressed the cultural
masculinity of leader stereotypes. Best known is the think
managerthink male paradigm, which was created by Schein
(1973). This method provides a direct test of the similarity of
leader stereotypes to male and female stereotypes. In these studies,
separate groups of participants rated a leader category (e.g., successful middle managers), women, or men on a large number of
gender-stereotypical traits. The researchers correlated the mean
ratings of managers or leaders with the mean ratings of men and,
separately, with the mean ratings of women. These correlations
represent the similarity of stereotypes of men and women to
cultural concepts of leadership. The think managerthink male
effect occurs when men and leaders are similar and women and
leaders are not similar. Although these studies typically provide
only these correlations and not information on the content of the
stereotypes of men, women, or leaders, the method yields a direct
test of the central assumptions of Eagly and Karaus (2002) role
congruity model and Heilmans (1983, 2001) lack-of-fit model as
applied to leader roles.
A second method, the agency communion paradigm, consists
of studies that examined the gender-stereotypical content of the
leader stereotype (Powell & Butterfield, 1979). In these studies,
participants rated a leader category (e.g., good manager) on separate masculine (i.e., agentic) and feminine (i.e., communal) gender stereotyping scales. Comparison of participants mean ratings
on the two scales determined whether the stereotype of leaders was
more masculine than feminine. By directly addressing the content
of leader roles, this method complements but differs from the
similarity comparisons yielded by the think managerthink male
paradigm.
A third method, the masculinityfemininity paradigm, consists
of studies that appeared in the research literature as a test of the
masculine versus feminine content of occupational stereotypes
(Shinar, 1975). A substantial minority of the occupations chosen
for these studies were leader roles (e.g., university president,
mayor). Participants rated each of these leader roles, among other
618
KOENIG, EAGLY, MITCHELL, AND RISTIKARI
occupational groups, on a single bipolar masculine versus feminine

rating scale. This paradigm became less popular subsequent to
critiques pointing out that bipolar masculinityfemininity scales do
not allow masculinity and femininity to vary independently but
force them to function as opposites (e.g., Constantinople, 1973).
Nonetheless, these data are informative in part because researchers
investigated various specific types of leaders rather than merely
leaders (or managers) in general. If the mean ratings of most
categories of leaders were on the masculine side of a masculinity
femininity scale, the studies would provide a conceptual replication of the basic tendency for leader roles to be stereotyped as
masculine.
Our project separately meta-analyzed studies in these three
paradigms because their distinctively different study designs and
measuring instruments precluded combination across the paradigms (see Borenstein, Hedges, Higgins, & Rothstein, 2009, pp.
359 363). The think managerthink male studies correlate (a)
ratings of men with ratings of managers and (b) ratings of women
with ratings of managers. In this think managerthink male paradigm, participants are randomly assigned to rate one of the three
groups (women, men, or leaders). In contrast, in the much simpler
agency communion and masculinityfemininity paradigms, studies obtain ratings only of leaders in general or specific leader roles
(and not women or men) on gender-stereotypical traits or scales.
Although the agency communion and masculinityfemininity
studies are similar in presenting only a leader group (or groups) to
participants, these two paradigms feature different types of measuring instruments. The masculinityfemininity studies use a single bipolar masculine versus feminine rating scale, whereas the
agency communion studies use two separate scales, one assessing
agency (or masculinity) and the other assessing communion (or
femininity). Therefore, in the masculinityfemininity paradigm,
the effect sizes are based on a point estimate (in relation to the
scale midpoint). In the agency communion paradigm, the effect
sizes are based on a comparison between ratings on the two scales.
For these reasons, the data are noncomparable across the three
paradigms and analyzed separately. Yet all three types of studies
test the correspondence between gender and leader stereotypes,
and the studies are methodologically quite homogenous within
each paradigm.
Variation in Stereotypes About Men, Women, and

Leaders
Eagly and Karau (2002) hypothesized that the incongruity between leader stereotypes and the female gender stereotype is not
fixed but varies with change in either stereotype. This metaanalysis examines several factors hypothesized to influence this
incongruity.
Change in Stereotypes Over Time

Cultural change over historical time is one of these factors,
given that stereotypes may have changed in a manner that reduces
womens role incongruity in relation to leadership. Although some
hints of the decreasing masculinity of leadership have appeared in
prior reviews (e.g., Duehr & Bono, 2006; Eagly & Sczesny, 2009;
Powell et al., 2002; Schein, 2001), generalizations have remained
uncertain. Temporal comparisons of relevant studies based on their
dates of publication, which extend back to 1973, allow our metaanalysis to clarify whether the cultural representation of leadership
has changed. A shift in an androgynous direction would ease
womens role incongruity problem in relation to leader roles.
Why might role incongruity have lessened? Organizational experts have often argued that definitions of good managerial practices have changed in response to features of the contemporary
organizational environment, such as fast social and technological
change and unprecedented complexity of organizations missions
and contexts (e.g., Avolio, 1999; Kanter, 1997; Lipman-Blumen,
2000). According to such analyses, these changed conditions compromise the efficacy of top-down command-and-control leadership
and foster democratic relationships, participatory decision-making,
delegation, and team-based leadership skills (e.g., Gergen, 2005;
Kanter, 1997; Lipman-Blumen, 2000; McCauley, 2004). Such
descriptions are manifestly less masculine than many traditional
models of good leadership.
Another possibility is that the increase of women leaders might
produce less masculine and more androgynous beliefs about leadership. Evidence of the increase of women leaders abounds, including growth over time in womens emergence as leaders in field
and laboratory studies of leader emergence in initially leaderless
groups (Eagly & Karau, 1991). In the United States, women now
constitute 25% of chief executives when all organizations are
considered and 43% of all employees in management, business,
and financial operations occupations (vs. 31% in 1983; U.S. Department of Labor, 2007, 2010b). Women have also become more
numerous in highly visible political leader roles, now constituting
17% of the U.S. Congress (vs. 2% in 1950; Center for American
Women and Politics, 2011) and 12% of governors (vs. 0% in 1950;
Center for American Women and Politics, 2010). Women have
increased in leadership roles in many other nations as well (see
European Commission, 2010; Inter-Parliamentary Union, 2010).
Evidence that the mere presence of more women leaders can
change perceptions of leader roles emerged in research on womens occupancy of the chief village councilor role in West Bengal
(Beaman, Chattopadhyay, Duflo, Pande, & Topalova, 2009). People in the villages that were mandated (vs. not mandated) by the
government to elect a woman for this leader role not only perceived their current women leaders and women leaders in general
as more effective but also reduced their implicit bias toward
associating men with leadership and elected more women to leadership positions in the next election. As additional evidence that
the presence of women leaders changes perceptions of leadership,
female college students with more women professors as role
models reduced their implicit associations of leadership qualities
with men and communal qualities with women (Dasgupta &
Asgari, 2004). Empirical research thus has demonstrated that an
increase in the number of women leaders can produce a more
androgynous concept of leadership and thereby reduce bias toward
current and potential women leaders.
What about change in gender stereotypes? If gender stereotypes
reflect the differing placements of men and women into social
roles (Wood & Eagly, 2010), womens increase in labor force
participation (to 61% vs. 33% in 1950; U.S. Department of Labor,
2010a) and in leader roles might predict change in the female
stereotype. However, women still perform the majority of domestic work (e.g., Bianchi, Robinson, & Milkie, 2006), and the majority of employed women have remained concentrated in tradi-
619
tional occupations. The six most common in the United States are
secretary and administrative assistant; registered nurse; elementary
and middle school teacher; cashier; retail salespersons; and nursing, psychiatric, and home health aides (U.S. Department of Labor,
2011). Therefore, it is not surprising that the partial reviews of
gender stereotyping conducted so far have not yielded evidence of
decreased stereotyping over time (e.g., Lueptow, Garovich-Szabo,
& Lueptow, 2001). Nonetheless, the masculinity of leader roles
could be changing without much change in stereotypes about men
and women in general.
In sum, leader stereotypes may have become less masculine
over time. If the change in leader stereotypes is related to increases
of women in leadership roles, then the number of women leaders
should be related to the masculinity of leadership. Because stereotype change presumably reflects the updating of impressions based
on new observations (e.g., Weber & Crocker, 1983), leader stereotypes at any one time point should correspond best to contemporaneous observations of women in leader roles.
Other Influences on the Masculinity of Leader Roles

A priori moderators. As Eagly and Karau (2002) argued,
men may have a more masculine leader stereotype than women do.
There is some existing evidence that men, more than women,
believe that good leaders have masculine qualities (e.g., Atwater,
Brett, Waldman, DiMare, & Hayden, 2004; Schein, 2001) and
manifest prejudice against female leaders (Eagly et al., 1992).
Such effects are understandable, given that men are less likely than
women to have experience with female managers (McTague,
Stainback, & Tomaskovic-Devey, 2009; Stainback &
Tomaskovic-Devey, 2009) and that mens group interest favors
retaining these roles for men. As a result, women leaders, particularly as newcomers entering male-dominated roles, can encounter
resistance, especially from men (Eagly & Carli, 2007). Our metaanalysis offers quantitative tests of the extent to which the incongruity between women and leaders is greater in men than women.
Role incongruity should also reflect organizational contexts and
the level of leader roles in organizational hierarchies (Eagly &
Karau, 2002). Leadership may be less masculine, for example, in
female-dominated fields such as elementary education, nursing, or
librarianship (U.S. Department of Labor, 2010b). Because these
fields are thought to require traditionally feminine skills, such as
warmth, compassion, and caring for others (Cejka & Eagly, 1999;
Glick, 1991), the characteristics that people associate with leadership roles in these occupations are likely to incorporate more
communal attributes. Also, consistent with the preponderance of
men in executive positions in many organizations (e.g., 84% of
corporate officers in Fortune 500 companies; Catalyst, 2010),
construals of leadership are likely to be more masculine for higher
status leader positions, thereby increasing role incongruity for
women. Our inclusion of studies examining various types of leadership positions allows comparisons of leader stereotypes between
different organizational contexts and levels of leadership within
organizations.
Comparisons of leader stereotypes across nations are also important, as Eagly and Karau (2002) also argued. Narrative reviewers have suggested that leader roles are less masculine in the
United States than in other nations (e.g., Schein, 2001), but there
has been no quantitative analysis across nations. Variation in the
percentages of women (vs. men) in leader roles could underlie any

national differences as well as temporal differences. The participation of women in leader roles and the overall status of women
are considerably greater in Western than Eastern nations (Hausmann, Tyson, & Zahidi, 2009). Thus, there is reason to believe that
role incongruity for women leaders is greater in Eastern than
Western nations.
In summary, consistent with role congruity theory, we predicted
that this meta-analysis would show a robust tendency for leader
roles to be perceived as masculine. However, based on this theory
and the other evidence outlined above, we expected that the
relationship between perceptions of leadership and masculinity
would be moderated by several factors. Masculinity of leadership
should decrease over time. In addition, men should have a more
masculine construal of leadership than women. Also, leadership
may be more masculine in domains with few women managers and
higher status leader roles. People in different nations may also
diverge in their construals of leadership, with the presence of fewer
women in leader roles associated with more masculine stereotypes.
These a priori moderators of the gender typing of leadership are
examined within this meta-analysis.
Exploratory Moderators. Several other variables were also
investigated as potential moderators of the masculinity of leadership on an exploratory basis. For example, because employment
generally entails some experience with women managers, older
participants might have a less masculine concept of leadership.
However, older peoples more traditional attitudes about gender
(e.g., Howell & Day, 2000) might foster a more masculine concept
of leadership. Therefore, we could not predict whether age would
be positively or negatively associated with the masculinity of
leadership but address the issue with an exploratory analysis.
Also, in research on gender, authors of one sex have occasionally obtained different findings than authors of the other sex (e.g.,
Eagly & Carli, 1981; Leaper, Anderson, & Sanders, 1998). In
addition, originators of research paradigms sometimes obtained
stronger data than other researchers (e.g., B. T. Johnson & Eagly,
1989; Wood, Lundgren, Ouellette, Busceme, & Blackstone, 1994).
Because originators stronger findings might partially be due to
their choice of stereotype measure in the case of the masculinity of
leadership, we also explored whether the differing measures used
in each paradigm were associated with different effects. Although
these variables have influenced other meta-analyses, we had no
specific reason to expect they would also influence the masculinity
of leadership, so we tested their effects as exploratory moderators.
Method
Sample of Studies and Criteria for Inclusion and
Exclusion
Three paradigms of research. The search located studies
that fit into the three different paradigms, whose defining characteristics are the following:
1. In the Schein (1973) think managerthink male paradigm,
participants rated a leader category, men (typically men in general), or women (typically women in general) on various traits
in a between-subjects design (with only two studies having a
within-subjects design). In most of these studies, the ratings were
completed using the Schein Descriptive Index, which consists of
620
92 traits, including many agentic or communal personality traits

(e.g., adventurous, submissive, aggressive, intuitive, ambitious,
modest, kind). Studies collected for this meta-analysis reported
internal consistency coefficients ranging from .71 to .92 for the
Schein Descriptive Index. Researchers reported intraclass correlations, computed across the traits, for relating the mean ratings of
leaders and men and of leaders and women.2 Studies were accepted even if they reported only a menleader or womenleader
correlation, but the great majority of studies reported both correlations.
2. In the Powell and Butterfield (1979) agency communion
paradigm, participants rated leaders (or managers) on two multiitem scales, typically the masculine and feminine scales of the
Bem Sex Role Inventory (Bem, 1974). One scale contains stereotypically masculine (i.e., agentic) personality traits, such as assertive, forceful, dominant, and competitive; the other scale contains
stereotypically feminine (i.e., communal) personality traits, such
as affectionate, compassionate, warm, and gentle. Studies collected for this meta-analysis reported internal consistency coefficients for the Bem Sex Role Inventory from .74 to .90 for the
agency scale and .79 to .90 for the communion scale. Researchers
usually reported means and standard deviations on the two scales
but sometimes presented item-level data that included a representation of agentic and communal traits or classified respondents
leader ratings into the four quadrants of a two-dimensional
agency communion space. The researchers other than Powell and
Butterfield whose studies fit this paradigm usually did not cite the
precedent of the Powell and Butterfield study, but nonetheless
reported participants ratings of a leader category on agency and
communion.
3. In the Shinar (1975) masculinityfemininity paradigm, participants rated one or more leader categories on a single bipolar
7-point scale that ranged from very masculine to very feminine.
Researchers reported means and standard deviations of these ratings and compared means to the midscale value, which represents
the gender-neutral or androgynous point on the scale. The metaanalysis includes only those occupations from each study that have
a clear leadership focus (most commonly managerial roles, such as
office manager).
The studies selected for all three of these paradigms presented
participants with a general leader category, such as managers or
executives, or with occupations or job titles denoting managerial
authority, such as personnel directors, head librarians, academic
administrators, military officers, or political office holders. For all
paradigms, data were excluded if they provided ratings of leader
behaviors rather than personality traits (e.g., Gutek & Stevens,
1979), of specific people in a leadership position (e.g., Petty &
Miles, 1976), or of leader groups identified by sex (e.g., successful female managers; Dodge, Gilroy, & Fenzel, 1995). Studies in
the agency communion paradigm were eliminated if the majority
of the items in their two scales did not pertain to agentic or
communal personality traits or if the two types of items were very
unequally represented (e.g., Buttner & Rosen, 1988).
The abstract of each identified document was evaluated by at
least one of the authors; if the study potentially fit the inclusion
criteria, the full document was obtained. In total, 78 documents
were rejected based on their abstracts and 134 documents were
rejected after reading the document. In addition, 11 documents met
the inclusion criteria but did not report the appropriate statistics to
calculate an effect size, and when contacted, the authors could not
supply the needed information (e.g., Couch & Sigler, 2001; Gerstner & Day, 1994).
Studies from any participant population or nation were included
if they fit the above criteria. When documents included data from
different nations or different participant samples within a nation,
they were treated as separate studies if the results were reported
separately. Some documents included data for more than one type
of leader category (e.g., company president, head librarian). If
separate groups of participants rated each category, the leader
categories were treated as separate studies. If the same participants
rated more than one leader category, these effect sizes were averaged prior to calculating study-level mean effect sizes but retained
as separate effect sizes for moderator analyses. In two documents
in the think managerthink male paradigm (Dodge et al., 1995;
Karau, 2005), separate groups of participants rated different types
of leaders. These ratings were correlated with the same ratings of
women and men, provided by two other groups of participants. In
our analyses, we treated these semi-independent leader conditions
as separate studies.
Some data qualified in more than one paradigm. For example,
we extracted only the leader condition of think managerthink
male studies for use in the agency communion paradigm if the
report contained item-level data that allowed us to separate agentic
and communal items into subscales (i.e., van Engen, 2006). Also,
parts of the same study that were published separately sometimes
qualified for different paradigms. For example, some authors presented intraclass correlation coefficients in one publication but
grouped the data into agentic and communal scales in another
(Fullagar, Sverke, Sverke, Sumer, & Slick, 2003; Sumer, 2006) or
presented the intraclass correlation coefficients in one publication
but the mean for a specific masculinefeminine bipolar scale in
another (Koch, Luft, & Kruse, 2005; Luft, 2003).
Search for studies.
Computer-based information searches
were conducted in the following databases: ABI/INFORM, Academic Press/Ideal, Business Source Elite, Proquest Digital Dissertations, Educational Resources Information Center (ERIC), Emerald Full Text, PAIS International, Proquest Business Databases,
PsycINFO, ScienceDirect, Sociological Abstracts, Web of Science, WilsonWeb, and Worldwide Political Science Abstracts.
In these searches, the keywords leader! (-s, -ship), manager!
(-s, -ial), educator, executive, candidate, public office, political
office, principal, or occupation were combined with (a) stereotyp!
(-e, -es, -ical), traits, characterization, attribute inventory, image,
2
The design of the studies in this paradigm appears similar to that of a

study by Broverman, Broverman, Clarkson, Rosenkrantz, and Vogel
(1970), whose participants rated a mature, healthy, socially competent
man, woman, or adult person. Broverman et al. found greater similarity
between an adult person and men than between an adult person and
women, but this finding has been criticized as artifactual (see Kelley &
Blashfield, 2008; Widiger & Settle, 1987, for details). However, the
similarity of the Broverman et al. study to the think managerthink male
studies is only superficial because Scheins (1973) method offers superior
item selection and statistical analysis. Specifically, (a) the typical think
managerthink male items (in the Schein Descriptive Index) are moderately balanced between agentic and communal qualities (see Duehr &
Bono, 2006), and (b) the statistical analysis with an intraclass correlation is
more appropriate to the data.
expectation, and perce! (-ption, -tive) or (b) gender terms, such as

man, men, woman, women, male, female, masculinity, femininity, sex,
sex role, sex-typing, gender, gender role, and androgy! (-ny, -nous).
Additional searches paired the two most common dependent measures (the Schein Descriptive Index and the Bem Sex Role Inventory) with manage! or leader!, think managerthink male, or
requisite management characteristics. Web of Science citation
searches were also performed for the seminal articles in each
paradigm (Powell & Butterfield, 1979; Schein, 1973; Shinar,
1975). All obtained studies reference sections were also searched
for relevant studies.
Several foreign psychological and academic databases were
searched with the limited keywords leader! (-s, -ship) or manager!
(-s, -ial) paired with (a) stereotyp! (-e, -es, -ical), traits, or similarities or (b) gender terms, such as masculinity, femininity, gender, and sex (translated, as appropriate). The foreign databases
were from Finland (University of Joensuu database, Forum of
Science database, Finnish Social Science Achieve, University of
Rovaniemi database, University of Tampere database); Germany
(Datenbank, PSYNDEX, PSYNDEXalert, PSYTKOM); Great
Britain and Ireland (Index to Theses); Norway (National Library of
Norway, BIBSYS Library); Spain (Psicodoc); and Sweden (Chalmers University of Technology database, the Center for Research
Libraries). Documents in languages other than English were translated by appropriately skilled individuals who assisted the two
coders.
To locate unpublished studies, messages asking for relevant data
were sent to the e-mail distribution lists of several organizations:
European Association of Social Psychology, European Association of Work and Organizational Psychology, Interamerican Society of Psychology, Society for the Psychology of Women, Social
Issues in Management Division of the Academy of Management,
Society for Personality and Social Psychology, and Society for the
Psychological Study of Social Issues. Data were also sought from
the originators of each of the paradigms, but they had no additional
data to offer. Finally, when the authors of this article gave several
talks on the meta-analysis in progress, they asked the audience to
contribute their own data or to provide leads about relevant sources
of data. Both unpublished and published studies meeting the inclusion criteria were included in the meta-analysis.
The initial search extended from the beginning of each paradigm
through the end of 2002. The search was then updated in June 2007
by consulting the databases that had yielded studies in 2002, with
the addition of Google and Google Scholar, and updated once
more in March 2009 with PsycINFO, Google, and Google Scholar.
Variables Representing Each Study in All Paradigms

The coded variables represent the a priori moderators as well as
other variables that we investigated on an exploratory basis. The
included studies, their codings, and effect sizes appear in Tables 1, 2,
and 3 for the think managerthink male, agency communion, and
masculinityfemininity paradigms, respectively.
A priori moderators. Year of publication (or year of data
collection for unpublished data) was recorded as well as the
percentage of male participants. Effect sizes were also calculated
separately by participant sex when possible. If data were not
reported separately for male and female participants, we contacted
the authors and requested this information.
621
For each paradigm, we recorded the exact description of the

leader group and coded its domain: (a) In the think managerthink
male paradigm, the leader domains were nearly all designated as
managerial, and those few not so designated were heterogeneous
(e.g., leader, educational administrator), so no moderator variable
was tested; (b) in the agency communion paradigm, the domains
were managerial (e.g., manager, middle-level manager), political
(e.g., mayor, state senator), or educational (e.g., elementary school
principal, school superintendent); and (c) in the masculinity
femininity paradigm, the domains, or economic sectors, of the
occupations were business (e.g., company president, marketing
manager), education (e.g., educational administrator, school principal), politics (mayor, politician), judicial (e.g., federal judge,
Supreme Court justice), arts (orchestra conductor, theatrical director), or other (boat captain, farm manager, park manager). For all
three paradigms, leader status was coded as high (defined as
prestigious political roles such as president or governor or organizational positions higher than middle manager such as upperlevel managers or executives) or moderate (e.g., managers,
middle-level managers, all other leader occupations).
Participant nationality was coded by the nation where the study
was conducted and classified as Western (United States, Great
Britain, Germany, Canada, New Zealand, Australia, Sweden,
South Africa, Netherlands, Denmark, and Portugal) or Eastern
(China, Japan, Hong Kong, Egypt, Saudi Arabia, and Turkey).
Other exploratory classifications (e.g., North American vs. other
Western vs. Eastern) did not improve prediction of the effect sizes.
To interpret cross-temporal and cross-national comparisons of studies, we imported data on the percentage of female managers, defined
as the percentage of managers who are women, based on the likely
time of data collection for the nations in which the studies were
conducted, defined as 2 years prior to studies dates of publication.
When available, these data came from the Human Development
Reports of the United Nations Development Programme (e.g., 2006)
and otherwise from the Yearbooks of Labour Statistics of the International Labour Organization (e.g., 1985). Aggregate indexes of the
status of women available from the United Nations Development
Programme or other sources were not appropriate for this metaanalysis because they are available only for more recent years.
Exploratory moderators. Average participant age was either
provided in each research report or estimated from sample characteristics. The distributions of the effect sizes across the types of
participant populations were as follows: (a) for the think manager
think male paradigm, 55% undergraduate students, 18% managers,
8% MBA students, 6% nonmanagerial employees, and 14% other
or mixed; (b) for the agency communion paradigm, 62% undergraduate students, 13% nonmanagerial employees, 9% managers,
9% MBA students, and 9% other or mixed; and (c) for the
masculinityfemininity paradigm, 86% undergraduate students
and 14% other or mixed.
The percentage of male authors was coded. Research group was
coded as originators of paradigm (Schein, Powell and Butterfield, or
Shinar) or other. Stereotype measure was coded based on the use of
the originators versus other measures: (a) In the think managerthink
male paradigm, Schein Descriptive Index or other (e.g., task vs.
person-orientation scales); (b) in the agency communion paradigm,
Bem Sex Role Inventory or other (e.g., initiating structure vs. consideration scales); and (c) in the masculinityfemininity paradigm,
always a masculinityfemininity 7-point scale.
Y
Y
1
1
1
Brenner et al. (1989)
Byler (2000)
Cohen-Kaner (1992)
Dion & Schuller (1990)

Dion & Schuller (1991)
Dodge et al. (1995)
Duehr & Bono (2003)
Duehr & Bono (2006)
Fernandes & Cabral-Cardoso

(2003)
Foster (1994)
Fullager et al. (2003)
Heilman et al. (1989)
Karau (2005)
Karau & Elsaid (2005)
Karau & Hansen (2005)
Kent (1984)
Y
Y
Y
Boyce & Herd (2003)
Sep.
by sexb
Samplea
Booysen & Nkomo (2006)
Report
1
1
1
Pub.
sourcec
100
100
100
50
50
50
50
50
33
67
% male
authors
2
2
1
Stereotype
measured
28
42
US
US
31
EG
SE
42
42
US
US
35
30
SE
US
32
TR
GB
32
42
US
PT
42
42
36
37
40
37
44
35
46
30
% female
mgr.
US
US
US/CA
US/CA
US
US
US
US
US
ZA
Natione
47
21
21
21
22
22
41
21
21
42
21
20
47
30
32
33
31
36
44
42
20
32
M age
SSA
SMM
SMM
SMM
CEO
SMM
SMM
SMM
SMM
SMM
SMM
SMM
SMM
SUM
SMM
1
1
1
Leader
statusg
SMM
SMM
SMM
SMM
EAD
SMM
SMO
SMM
Leader
rolef
Table 1
All Included Studies (k ! 51) With Moderator Values and Effect Sizes for the Think ManagerThink Male Paradigm
391
404
508
501
407
395
83
85
119
119
53
43
48
48
48
48
76
74
178
169
197
203
199
199
160
160
536
536
182
182
78
78
112
104
113
105
217
220
371
357
47
47
61
61
240
247
n
69
69
82
82
72
69
82
78
39
39
45
48
59
59
59
59
39
31
60
57
30
33
46
46
50
50
58
58
27
27
100
100
59
63
58
63
58
59
75
73
55
57
51
56
47
52
% male
part.
92
92
92
92
92
92
92
92
92
92
51
22
92
92
92
92
92
92
92
92
92
92
42
42
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
Item n
W
M
W
M
W
M
W
M
W
M
M
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
Leader
sim.h
.46
.70
".03
.39
.15
.68
.45
.23
.26
.53
.58
.66
.15
.50
.02
.59
.54
.56
.67
.56
.28
.43
".01
.40
.11
.46
.21
.58
.15
.68
".24
.54
.01
.39
.19
.93
.31
.40
".24
.43
.44
.33
.28
.39
.58
.59
ICCi
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.02
0.05
0.01
0.25
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.03
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
Variance
622
Y
Y
Marmenout (2008)
Massengill & DiMarco

(1979)
Mook (2005)
Neergaard et al. (2007)
Norris & Wylie (1995)
Orser (1994)j
Paradine et al. (1995)
Ryan et al. (in press)
Sauers et al. (2002)
Schein (1973)
Schein (1975)
Schein & Davidson

(1993)k
Schein & Mueller

(1992)
Schein et al. (1989)
Schein et al. (1996)
Sczesny (2003a)
Luft (2003)
Sep.
by sexb
Samplea
Kunkel et al. (2003)
Report
Table 1 (continued)
Pub.
sourcec
50
33
50
67
25
66
33
100
33
% male
authors
Stereotype
measured
27
12
CN
DE
JP
35
19
DE
US
32
32
GB
GB
32
17
18
38
37
GB
US
US
NZ
AU
40
41
CA
US
41
40
25
26
22
31
27
46
% female
mgr.
CA
US/CA
DK
NL
US
SA
DE
US
Natione
24
21
21
21
24
21
21
21
44
43
21
27
22
29
21
21
21
43
39
44
21
28
21
M age
SMM
SMM
SMM
SMM
SMM
SMM
SMM
SMM
SMM
SMM
MSC
MSC
SMM
SM
SM
SMM
MSC
SM
SM
SMM
CEO
Leader
rolef
Leader
statusg
41
42
202
202
144
171
89
83
85
96
119
113
613
605
198
198
198
198
204
213
35
36
91
75
212
212
193
212
113
115
152
152
101
101
101
101
297
297
152
152
195
210
186
183
73
73
n
46
46
47
47
40
37
60
57
48
45
85
82
48
48
0
0
50
50
55
54
38
51
22
25
48
48
100
100
0
0
54
54
48
48
48
48
63
63
64
64
66
64
56
53
58
58
% male
part.
92
92
25
25
92
92
92
92
92
92
92
92
24
24
92
92
92
92
40
40
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
40
40
Item n
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
W
M
Leader
sim.h
Variance
.18
0.01
.80
0.01
".27
0.04
.68
0.04
.18
0.01
.61
0.01
.15
0.01
.64
0.01
.24
0.01
.58
0.01
.44
0.01
.72
0.01
.44
0.05
.81
0.05
.31
0.01
.46
0.01
.18
0.01
.49
0.01
.15
0.03
.76
0.03
.36
0.01
.72
0.01
.46
0.01
.79
0.01
.43
0.01
.71
0.01
.06
0.01
.62
0.01
.30
0.01
.54
0.01
.19
0.01
.52
0.01
.17
0.01
.64
0.01
.17
0.01
.64
0.01
.05
0.01
.71
0.01
.23
0.01
.64
0.01
".06
0.01
.60
0.01
.10
0.01
.91
0.01
.65
0.03
.76
0.03
(table continues)
ICCi
623
624
1
SMM
19
25
HK
2
50
Y
1
Yim & Bond (2002)
1
SMM
22
26
NL
1
0
2
1
van Engen (2006)
1
SM
38
27
US
1
0
2
Y
1
Sylvan (1983)
Note. In each sample, separate groups of participants rated a leader category, women, or men on gender-stereotypical traits. The effect sizes are the intraclass correlations between the mean ratings
of managers (or leaders) and (a) the mean ratings of men or (b) the mean ratings of women across all the traits. Sep. ! separated; pub. ! publication; mgr. ! managers; part. ! participants; sim. !
similarity; ICC ! intraclass correlation coefficient.
a
The number of different samples (e.g., from different populations or countries) reported within the article. b Separated samples by sex: N ! no; Y ! yes; M ! sample was all male; F ! sample
was all female. c Publication source: 1 ! published; 2 ! unpublished. d 1 ! Schein Descriptive Index; 2 ! other. e AU ! Australia; CA ! Canada; CN ! China; DK ! Denmark; EG ! Egypt;
DE ! Germany; GB ! Great Britain; HK ! Hong Kong; JP ! Japan; NL ! Netherlands; NZ ! New Zealand; PT ! Portugal; SA ! Saudi Arabia; ZA ! South Africa; SE ! Sweden; TR ! Turkey;
US ! United States; X ! included data from several countries. f EAD ! effective athletic directors; L ! leadership; M ! managers; MSC ! managers of successful companies; CEO ! successful
CEOs; SM ! successful managers; SMM ! successful middle managers; SMO ! successful military officers; SSA ! successful school administrators; SUM ! successful upper level managers. g 1 !
moderate; 2 ! high. h M ! menleaders similarity; W ! womenleaders similarity. i Effect size (ICC) computed using a one-way single rater random effects model, related the mean ratings of
the leader role to the ratings of men or women. These correlations represent the similarity of stereotypes of men or women to stereotypes of leaders; higher correlations indicate a greater
similarity. j The sample with 50% male participants was composed of business students, and the sample with 0% male participants was composed of nonbusiness students. k The sample with 54%
male participants was from North Great Britain, and the sample with 48% male participants was from South Great Britain.
0.03
0.03
0.01
0.01
0.01
0.01
0.03
0.03
.87
.87
.11
.48
.29
.38
.44
.72
W
M
W
M
W
M
W
M
40
40
92
92
92
92
32
32
Y
Sczesny (2003b)
DE
27
45
60
64
60
63
70
31
142
142
58
70
53
54
23
0
49
49
Leader
sim.h
Report
Table 1 (continued)
Samplea
Sep.
by sexb
Pub.
sourcec
% male
authors
Stereotype
measured
Natione
% female
mgr.
M age
Leader
rolef
Leader
statusg
% male
part.
Item n
ICCi
Variance
Lastly, the source of publication was coded as published or

unpublished (including dissertations, masters theses, and other
unpublished documents) to provide one estimate of possible publication bias (Sutton, 2009; see the subsection Publication Bias).
Two of the authors, who had successfully completed a course on
meta-analysis, independently coded the studies, with high agreement across the variables. The mean kappas for intercoder agreement were .90, .90, and .97 for the think managerthink male,
agency communion, and masculinityfemininity paradigms, respectively. Disagreements were resolved by discussion.
Computation of Effect Sizes and Data Analysis

Effect sizes were calculated with a hand calculator or DSTAT
software and then entered into Comprehensive Meta-Analysis
(Version 2.2.050) and Statistical Package for the Social Sciences
(SPSS). Subgroup analyses were conducted using Comprehensive
Meta-Analysis, and meta-regressions were conducted with SPSS
macros provided by D. Wilson (see http://mason.gmu.edu/
#dwilsonb/ma.html). For all analyses, p values less than .05 were
considered significant and values between .05 and .10 were defined as marginal.
Calculation of effect sizes. In the think managerthink male
paradigm, researchers reported intraclass correlation coefficients
(ICCs) for the menleaders and womenleaders relationships. In
all cases in which ICCs were not reported, the researchers provided
additional information that allowed us to produce intraclass correlations. The ICCs used in the primary data reports were computed using a one-way, single-rater, random effects model, which
assesses absolute agreement among measurements (Case 1 in
McGraw & Wong, 1996). However, the more appropriate ICC
would have been the 2-way ICC(A,1), which is also a measure of
absolute agreement but which takes the fixed column factor (leaders vs. women; leaders vs. men) into account. Thus, the ICC
provided by researchers in the primary studies would be biased
downward somewhat but quite close to the calculations for
ICC(A,1) (K. O. McGraw, personal communication, January 8,
2003). We used the one-way, single-rater, random effects ICC
because it was available for all data sets, providing a common
metric for the studies. The menleaders and womenleaders similarities, or ICCs, were analyzed separately as Fishers Z, using the
conversion .5 * log({1 $ [(k " 1) * r]}/(1 " r)), where k is the
number of observations made on each object of measurement, and
given an inverse variance within-study weight of 1/{k/[2 * (n "
2)(k " 1)]}, where n is the number of items (see formulas in
Appendix B of McGraw & Wong, 1996). The Zs were transformed
back to ICCs for presentation of the results.3 In the random-effects
models, the variances in these study weighting terms consisted of
the sum of the within-study variance and the between-studies
variance (see Borenstein et al., 2009, p. 72).
In the agency communion paradigm, researchers reported means
and standard deviations separately on the agentic and communal
scales, allowing the computation of a d effect size comparing the
3
Because the womenleaders and menleaders correlations are dependent in that they share the data for leaders, we did not compare the two
effects statistically. These types of comparisons would require information
about the correlation between male and female targets, which is not
available in the primary studies.
N
Y
Y
Y
Y
N
Y
Samplea
1
1
1
1
2
1
2
Report/leader role
Butterfield & Powell (1981)

Good president
Cann & Siegfried (1987)
Manager
Huddy & Terkildsen (1993)
Good Congress member
Good local council member
Good mayor
Good president
Average of roles
Inderlied & Powell (1979)
Good manager
Good manager
Good manager
Good manager
Maier (1993)
Typical manager
Average of roles
Ideal manager
Marongiu & Ekehammar (2000)
Manager
Powell (1992)
Good manager
Typical American manager
Powell & Butterfield (1979)
Good manager
Good manager
Good manager
Good president of the U.S.
Good vice-president
Good manager
Good manager
Powell et al. (2002)
Good manager
Good manager
Powell & Kido (1994)
Good manager
Typical Japanese manager
Rosenwasser & Dean (1989)
City council member
Governor
Mayor
Sep. by
sexb
Pub.
sourcec
10
67
100
100
100
100
100
50
100
50
100
100
% male
authors
Stereotype
measured
US
JP
US
US
US
US
US
US
SE
US
US
US
US
US
Natione
35
45
35
33
28
22
37
27
37
22
37
33
25
% female
mgr.
Table 2
All Included Studies (k ! 48) With Moderator Values and Effect Sizes for the AgencyCommunion Paradigm
21
21
21
32
21
20
27
20
20
20
27
41
21
43
37
28
22
20
21
21
20
M age
2
2
2
3
3
3
3
3
3
2
2
3
3
3
3
3
3
3
3
3
3
2
2
2
2
Leader
domainf
1
2
2
1
1
1
1
1
1
2
2
1
1
1
1
1
1
1
1
1
1
2
1
2
2
Leader
statusg
14
14
14
88
86
123
206
199
126
50
44
627
110
574
87
79
87
60
59
41
51
154
259
133
149
153
139
110
128
54
54
54
77
23
65
57
43
60
46
46
62
82
70
47
44
48
50
50
63
0
69
58
54
50
51
56
53
59
% male
part.
0.03
0.02
0.13
0.04
0.15
0.23
0.05
0.05
0.01
0.20
0.05
0.03
0.03
0.02
0.04
0.05
0.04
0.09
0.06
0.02
0.01
0.02
0.01
0.01
0.02
0.02
0.02
0.03
Variance
2.44
0.25
2.66
0.27
2.31
0.24
(table continues)
"1.18
0.27
1.49
1.08
2.56
1.97
1.53
0.81
2.47
2.35
1.98
1.46
1.63
0.65
0.72
1.79
1.25
2.65
2.14
1.41
1.22
0.48
0.20
0.75
1.00
0.61
0.79
2.36
gh
625
Y
Y
Y
Y
1
1
1
1
Sep. by
sexb
Samplea
Pub.
sourcec
% male
authors
Stereotype
measured
US
NL
GB
LU
NL
TR
NL
DE
Natione
35
27
33
26
26
19
% female
mgr.
50
22
43
43
22
21
36
23
45
M age
1
1
1
1
1
3
3
2
1
2
2
2
2
Leader
domainf
1
1
1
2
1
1
1
2
1
1
1
2
2
Leader
statusg
54
54
54
54
54
139
363
66
40
369
3229
109
54
14
14
14
14
14
14
43
43
43
43
43
52
33
50
28
55
67
64
100
54
54
54
54
54
54
% male
part.
1.00
0.76
0.28
0.34
1.20
0.71
1.19
1.71
1.02
1.00
1.29
1.07
3.11
1.84
1.40
1.01
2.40
1.59
2.55
1.98
gh
0.04
0.04
0.04
0.04
0.04
0.04
0.02
0.01
0.03
0.06
0.01
0.001
0.25
0.23
0.18
0.16
0.25
0.19
0.26
0.21
Variance
Note. In each sample, participants rated a leader role on separate masculine (i.e., agentic) and feminine (i.e., communal) gender stereotyping scales. Comparison of participants mean ratings on the
two scales determined whether the stereotype of leaders was more masculine than feminine. The effect size for the average of the roles is presented for studies using a within-subjects design. Sep. !
separated; pub. ! publication; mgr. ! managers; part. ! participants.
a
The number of different samples (e.g., from different populations or countries) reported within the article. Some studies with multiple samples reported participant characteristics across all
samples. b Separated samples by sex: N ! no; Y ! yes. c Publication source: 1 ! published; 2 ! unpublished. d 1 ! Bem Sex Role Inventory; 2 ! other. e DE ! Germany; GB ! Great Britain;
JP ! Japan; LU ! Luxembourg; NL ! Netherlands; SE ! Sweden; TR ! Turkey; US ! United States. f 1 ! educational; 2 ! political; 3 ! managerial. g 1 ! moderate; 2 ! high. h Effect
size (g) computed by subtracting the mean rating on the feminine scale from the mean rating on the masculine scale and dividing by the pooled standard deviation, adjusted for small sample bias. If
only frequencies or percentages in a 2 % 2 table created by splitting the scales at the medians were reported, agency and communion were treated as dichotomous, and g was estimated from dCox. Larger
gs indicate a more masculine concept of leadership.
President
School board member
State representative
State senator
U.S. representative
U.S. senator
Rustemeyer & Thrien (1989)
Good manager
Good manager
Stoker (2007)
Ideal manager
Sumer (2006)
Successful middle manager
van Engen (2006)
Successful middle manager
Vinnicombe & Cames (1998)
Successful manager at their
bank
Vinnicombe & Singh (2002)
Successful managers who
reached the top team
Willemsen (2002)
Successful manager at their
bank
Williams (1989)
Academic dean of students
Assistant principal
Elementary school principal
School superintendent
Secondary school principal
Average of roles
Report/leader role
Table 2 (continued)
626
Samplea
1
Report/leader role
Beggs & Doolittle (1993)

Boat captain
Company president
Dean/educational administrator
Director, labor relations
Director, operations
Farm manager
Federal judge
Head librarian
Manager, credit union
Manager, procurement services
Managing editor
Mayor
Motel manager
Office manager
Orchestra conductor
Park manager
Personnel director
Politician
Public relations director
Sales manager
Sales president
School principal
Theatrical director
University president
U.S. Supreme Court justice
Average of roles
Gatton et al. (1999)
Accounting manager
Computer operations manager
Data processing manager
Engineering manager
Financial manager
Human resources manager
Marketing manager
Office manager
Plant manager
Quality control manager
R & D director
Sales manager
Average of roles
Koch et al. (2005)
Leadership
Sep. by
sexb
33
% male
authors
DE
US
US
Nationc
27
44
37
% female
mgr.
28
21
21
M age
2
2
2
2
2
2
2
2
2
2
2
2
6
2
1
2
2
6
5
1
2
2
6
4
2
2
3
6
2
4
2
2
2
1
3
1
5
Leader
domaind
Table 3
All Included Studies (k ! 101) With Moderator Values and Effect Sizes for the MasculinityFemininity Paradigm
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
2
1
Leader
statuse
202
175
175
175
175
175
175
175
175
175
175
175
175
139
138
140
141
141
140
141
141
141
141
139
141
141
141
140
141
139
140
141
140
140
141
141
140
139
47
47
47
47
47
47
47
47
47
47
47
47
47
50
49
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
51
50
50
50
50
50
50
50
% male part.
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
Variance
0.63
0.005
(table continues)
0.84
0.67
"0.16
1.46
0.76
"0.31
0.20
0.27
1.77
0.84
0.45
0.38
0.60
2.44
1.22
0.61
0.88
1.11
1.64
1.46
"1.66
0.87
0.72
0.85
1.40
0.45
0.26
1.35
1.10
0.30
1.40
0.06
0.60
1.35
0.92
0.01
1.44
1.51
0.89
gf
627
Boat captain
Company president
Farm manager
Federal judge
Head librarian
Managing editor
Mayor
Motel manager
Orchestra conductor
Park manager
Personnel director
Politician
Sales manager
Sales president
School principal
Supreme Court justice
Mahoney & Blake (1987)

Farm manager
Financial manager
Sales manager
Average of roles
Munoz Sastre et al. (2000)
Boat captain
Company president
Farm manager
Federal judge
Head librarian
Managing editor
Mayor
Motel manager
Orchestra conductor
Park manager
Personnel director
Politician
Sales manager
Sales president
School principal
Supreme Court justice
Theatrical director
Average of roles
Sep. by
sexb
Samplea
Report/leader role
Table 3 (continued)
20
100
% male
authors
32
ES
33
% female
mgr.
FR
US
Nationc
21
21
26
M age
6
2
1
6
5
1
2
4
2
3
6
2
4
2
2
2
1
5
6
2
1
6
5
1
2
4
2
3
6
2
4
2
2
2
1
5
3
1
6
2
2
Leader
domaind
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
2
1
1
1
Leader
statuse
126
126
126
126
126
126
126
126
126
126
126
126
126
126
126
126
126
126
234
234
234
234
234
234
234
234
234
234
234
234
234
234
234
234
234
234
234
234
29
34
29
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
52
50
52
% male part.
2.03
0.75
0.74
1.25
0.59
"0.09
0.74
0.61
0.68
1.30
0.78
0.70
0.62
0.07
0.51
0.45
0.40
0.88
2.35
0.98
0.92
1.97
0.53
"1.03
0.52
0.78
0.80
1.22
0.93
0.49
1.15
0.28
0.30
0.72
0.07
0.60
0.53
0.87
0.75
1.52
0.48
0.58
0.86
gf
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.036
0.030
0.035
0.033
Variance
628
Samplea
Sep. by
sexb
% male
authors
US
Nationc
18
% female
mgr.
21
M age
6
2
1
6
5
1
2
4
2
3
6
2
4
2
2
2
1
3
1
5
3
1
Leader
domaind
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
2
1
1
2
Leader
statuse
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
126
126
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
41
41
% male part.
3.91
9.06
0.99
2.09
1.76
"1.59
0.68
1.73
1.08
2.73
3.03
0.50
0.96
0.68
1.13
1.62
0.67
1.09
3.94
7.08
2.16
0.32
0.57
0.70
gf
0.009
0.011
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.009
0.009
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.009
0.010
0.009
0.008
0.008
0.008
Variance
Note. In each sample, participants rated leader roles on a single bipolar masculinityfemininity scale. Comparison of these ratings to the midpoint of the scale determined whether the stereotype of
leaders was masculine or feminine. The effect size for the average of the roles is presented for studies using a within-subjects design. All studies were published and used a 17 masculinefeminine
rating scale. Sep. ! separated; mgr. ! managers; part. ! participants.
a
The number of different samples (e.g., from different populations or countries) reported within the article. b Separated samples by sex: N ! no; Y ! yes. c FR ! France; DE ! Germany; ES !
Spain; US ! United States. d 1 ! education; 2 ! business; 3 ! arts; 4 ! political; 5 ! judicial; 6 ! other. e 1 ! moderate; 2 ! high. f Effect size (g) computed by subtracting the midpoint of
the scale from the leader rating, divided by the standard deviation and corrected for small sample bias.
Theatrical director
Average of roles
Shinar (1975)
Boat captain
Company president
Farm manager
Federal judge
Head librarian
Managing editor
Mayor
Motel manager
Orchestra conductor
Park manager
Personnel director
Politician
Sales manager
Sales president
School principal
Theatrical director
U.S. Supreme Court justice
Average of roles
Report/leader role
Table 3 (continued)
629
630
ratings on the two scales: (M1 " M2)/sp. The effect sizes were
converted to g with the correction for small sample bias: 1" [3/
(4N " 9)] (Borenstein et al., 2009). Some authors split their sample at
the median on both scales and reported the frequencies or percentages
in each quadrant of the resulting 2 % 2 table. If only this report was
available, agency and communion were treated as dichotomous, and
g was estimated from dCox, which is a logistic transformation of the
odds-ratio (Sanchez-Meca, Marn-Martnez, & Chancon-Moscoso,
2003, Formula 18). The within-study weighting term was the conventional inverse variance for standardized comparisons of means
(Lipsey & Wilson, 2001, p. 72) or dCox (Sanchez-Meca et al., 2003,
Formula 19), with the random-effects models also incorporating the
between-studies variances in the study weight.
In the masculinityfemininity paradigm, authors reported means
and standard deviations on a 7-point masculinityfemininity scale,
which yielded a d as a comparison with the scales midpoint [(M "
midpoint)/SD]. The effect sizes were then corrected for small
sample bias with the formula 1 " [3/(4N " 5)] to create a g
(Borenstein et al., 2009). The within-study weighting term for
these effect sizes was (1/n) $ {(d * d)/[2n(n " 1)]} (B. Becker,
personal communication, June 19, 2008), and the random-effects
models included the between-studies variance in the study weight.
Mean and distribution of effect sizes. Within each paradigm, the presentation first considers the mean weighted effect size
(with within-subjects effect sizes combined prior to averaging),
calculated by a random-effects model because the studies effect
sizes were not assumed to be consistent with a single underlying
mean value. We tested for outliers, as defined by more than 1.5
times the interquartile range beyond the upper quartile (i.e., the
upper inner fence, Tukey, 1977). We used Cohens (1988) benchmarks for d and r to describe the magnitude of the effect sizes g
and ICC, under which a g of .20 or an ICC of .10 is considered
small, a g of .50 or an ICC of .30 is considered medium, and a g
of .80 or an ICC of .50 is considered large. Along with the overall
mean, we present several distributional statistics recommended by
Borenstein et al. (2009): (a) T (or tau, the estimated standard
deviation of the true effect sizes); (b) 95% confidence interval (CI;
a measure of the accuracy of the mean; 95% of mean effect sizes
would fall inside this interval); (c) 90% prediction interval (PI; a
measure of the dispersion of effect sizes; 90% of true effects in
new studies with a sample selected at random would fall inside this
interval); (d) Q (a measure of uncertainty, or whether heterogeneity is genuine); and (e) I2 (a measure of the magnitude of heterogeneity, defined as the proportion of the observed dispersion that
is real rather than spurious on a 1 to 100% scale). An I2 near zero
indicates that almost all of the observed variance in the effects is
spurious and that there is no variance to explain, whereas a large
I2 indicates that investigation of the reasons for this variance is
warranted.
Publication bias. Analyses of the distribution of effect sizes
checked for potential biases in publication or our retrieval of
studies (see Borenstein et al., 2009). We first examined the funnel
plot of the effect sizes plotted by the standard error and assessed
whether Eggers test of the plots asymmetry was significant. We
next implemented the trim-and-fill procedure, which estimates the
number of studies that should be removed to create a more symmetric funnel plot. This procedure assesses the impact of removing
these studies on the mean effect size, then fills these studies back
in and imputes a mirror image for each such study to correct the
variance. In addition, a small studies analysis using cumulative

meta-analysis was used to calculate the mean effect as each study
was added, starting with those with the largest sample sizes. This
analysis determines whether the mean effect size estimate has
stabilized with the larger studies and does not shift when smaller
studies are added (see Borenstein et al., 2009). Finally, a mixedeffects subgroup analysis tested whether published and unpublished studies differed in their mean effect size.
Moderator analyses.
We present subgroup analyses and
meta-regressions to examine whether moderators accounted for
variability among effect sizes (using each effect size as a separate
sample). For each subgroup analysis, we present the overall mean
and tau for each subgroup. Within each paradigm, we first examined the effect sizes as moderated by the theory-relevant, a priori
variables (i.e., publication year, participant sex, leader domain,
leader status, participant nationality, and percentage of female
managers) and then as moderated by the other, exploratory variables. In each paradigm, all coded variables were tested as single
predictors. However, although the results for all a priori variables
are presented for each paradigm, only exploratory variables that
produced significant moderation as single predictors are reported.
The subgroup analyses used mixed-effects models because these
subgroups involve specific grouping (such as males and females)
that are fixed across studies (Borenstein et al., 2009; Lipsey &
Wilson, 2001). These mixed models consisted of a random-effects
model within subgroups and a fixed-effect model across subgroups, with appropriate inverse variance weights (Borenstein et
al., 2009). Meta-regressions were calculated as random-effects
models, with appropriate inverse variance weights (Borenstein et
al., 2009). These calculations were simple meta-regressions for
each centered continuous moderator (e.g., year of publication) and
a multiple meta-regression for the moderators that were individually significant in the subgroup analyses or simple metaregressions. The categorical moderators were recoded into two
groups for entry into each multiple meta-regression based on
maximizing the differences between the subgroup categories, as
determined by the single-variable moderator analyses. The
random- and mixed-effects models were calculated using the
method of moments (Borenstein et al., 2009).
Results
Think ManagerThink Male Paradigm
Mean and distribution of effect sizes. Table 1 lists the 40
studies with their characteristics and 51 effect sizes. A more
masculine stereotype is indicated by (a) a greater menleaders
similarity, as indexed by a larger menleaders ICC; and (b) a lesser
womenleaders similarity, as indexed by a smaller womenleaders
ICC. The weighted mean effect for the 51 menleaders ICCs was
.62 (T ! 0.26; 95% CI [.57, .66]; 90% PI [.27, .82]). There was a
large amount of variability within the effect sizes, Q(50) ! 321.65,
p & .001; I2 ! 84.46. The weighted mean effect for the 49
womenleaders ICCs was .25 (T ! 0.22; 95% CI [.19, .32]; 90%
PI [".12, .56]). There was also a large amount of variability within
these effect sizes, Q(48) ! 241.77, p & .001; I2 ! 80.15. Thus,
consistent with an overall think managerthink male effect, the
menleaders similarity was large and the womenleaders similarity was small. Although the variability in each of these two types
of effect sizes was large, as indicated by T and the confidence and

prediction intervals, a high proportion of the variability reflected
real differences, as indicated by I2.4
Publication bias. Analyses checked for potential biases in
publication or our retrieval of studies for both the womenleaders
similarity and the menleaders similarity. Eggers test of the plots
asymmetry was not significant for either of the two sets of effect
sizes, suggesting no asymmetry. For the menleaders similarity,
under the assumption that biases would hinder the publication of
studies with small effects, the trim-and-fill procedure suggested
that no studies be trimmed. For the womenleaders similarity,
under the assumption that biases would hinder the publication of
studies with large effects (because they would be contrary to the
hypothesis of low similarity between women and leaders), the
trim-and-fill procedure suggested that four studies with small
effect sizes should be trimmed to make the plot more symmetric,
increasing the mean effect size from .25 to .29. Computing these
analyses with the effect size weights based on the number of
participants rather than the number of items also revealed no
significant asymmetry in the funnel plots, and no studies were
trimmed. In addition, using either the number of items or the
number of participants to order the data, cumulative meta-analyses
from large to small studies showed very little movement of the
effect size with the addition of small studies. Finally, a mixedeffects subgroup analysis comparing published and unpublished
studies yielded no difference for the womenleaders similarity,
QB(1) ! 0.28, p ! .60, or the menleaders similarity, QB(1) !
1.00, p ! .32. Thus, overall, there appears to be little publication
bias, and the bias that may exist would only slightly increase the
womenleaders similarity.
Analyses for each moderator. These analyses report the
impact of each moderator on the womenleaders similarity and the
menleaders similarity, as shown in Table 4 for the subgroup
analyses and Table 5 for the simple meta-regressions. Models are
presented for both womenleaders and menleaders effect sizes
for each theory-relevant variable, but they are presented for other
exploratory variables only when the model yielded a p value of .05
or less for either type of similarity.
Publication year (ranging from 1973 to 2010) was marginally
related to the womenleaders similarity, with earlier years associated with a weaker womenleaders similarity, but was not significantly related to the menleaders similarity. Male (vs. female)
participants were associated with a weaker womenleaders similarity, but this variable was not significantly related to the men
leaders similarity. With participant samples not divided by sex,
greater percentages of male participants were associated with a
marginally weaker womenleaders similarity but not significantly
related to the menleaders similarity.5
High-status (vs. moderate-status) leaders were associated with a
stronger menleaders similarity but not significantly related to the
womenleaders similarity. Eastern (vs. Western) participant nationalities were associated with a marginally weaker women
leaders similarity but not significantly related to the menleaders
similarity. Smaller percentages of female managers in participants
nations were not significantly related to the womenleaders or
menleaders similarity.
Among the exploratory moderators, older participant age was
associated with a stronger womenleaders similarity but not significantly related to the menleaders similarity. Greater percent-
631
ages of male authors were associated with a weaker women

leaders similarity but not significantly related to the menleaders
similarity.
Multiple meta-regression analyses. Given some confounding between the predictors, it is important to examine each moderator controlled for the effects of the other moderators. Table 6
shows models that regressed both sets of effect sizes onto the
predictors that were at least marginally significant in any of the
subgroup or simple meta-regression models. However, this analysis omitted participant sex because it was created by splitting
studies participant samples. In the resulting model, the women
leaders similarity was stronger in later publication years, with a
lower percentage of male participants, and with older participants.
The menleaders similarity was stronger for higher status leaders.
AgencyCommunion Paradigm
Mean and distribution of effect sizes. Table 2 lists the 22
studies with their characteristics and 48 effect sizes. Larger gs
indicate a more masculine stereotype, as defined by the agency
minus communion difference. One sample (Powell & Kido, 1994;
ratings of good manager) was removed from further analyses as
an outlier (g ! "1.18). The weighted mean effect for the 39
remaining samples (merging the data within each of the three
studies that used within-subjects designs) was g ! 1.55 (T ! 0.55;
95% CI [1.36, 1.75]; 90% PI [0.61, 2.49]). Thus, the overall effect
was strongly in the masculine direction, with participants rating
leader groups (e.g., managers) as higher in agentic than communal
traits, with the lower bounds of the confidence and prediction
intervals remaining above zero. Although there was a large amount
of variability in the effect sizes, Q(38) ! 615.79, p & .001, I2 was
93.83, showing that a high proportion of the variability reflected
real differences, not random error.6
Publication bias. The funnel plot of the effect sizes plotted by
the standard error was asymmetric by Eggers test of asymmetry, with
smaller studies clustering to the right (masculine direction) of the
mean effect size. The trim-and-fill procedure suggested that 14 studies
with large effect sizes should be trimmed to make the plot more
symmetric, decreasing the mean effect size from 1.55 to 1.15. A
cumulative meta-analysis from large to small studies indicated
very little change in the overall effect when the smallest studies
4
We also recomputed the analyses using the number of participants as
the weight with generally similar results. The overall effects and associated
statistics were nearly identical, although the Q and I2 were larger for both
the womenleaders and menleaders similarities. Subgroup analyses and
meta-regressions produced findings similar to those obtained with item n
used to compute the weights.
5
Schein (2001; Schein, Mueller, & Jacobson, 1989) maintained that the
think managerthink male effect has decreased over time only for female
participants. However, although the womenleaders similarity was weaker
for female than male participants and increased over time, there was no
evidence of greater increase for female than male participants.
6
The weighted mean effect including the outlier was 1.49 (T ! 0.64;
95% CI [1.27, 1.71]; 90% PI [0.40, 2.58]), Q(39) ! 835.88, p & .001; I2 !
95.33. We also computed the same moderator analyses including the
outlier. The effects were similar to those reported here, although the leader
domain model was only marginal and the models for percentage of male
participants, percentage of male authors, and research group were nonsignificant.
632
Table 4
Subgroup Analyses for the Think ManagerThink Male Paradigm
Womenleaders similarity
Variable and class
QB
Participant sex
Male
Female
Leader status
Moderate
High
Participant nationality
Western
Eastern
30.38
&.001
0.82
2.88
Menleaders similarity
ICC
47
47
.11
.37
0.20
0.24
46
3
.26
.13
0.23
0.00
42
6
.27
.09
0.22
0.18
.37
QB
1.43
.23
8.72
.09
0.95
ICC
48
49
.63
.58
0.26
0.27
48
3
.60
.82
0.22
0.48
44
6
.60
.68
0.24
0.38
.003
.33
Note. QB ! between-classes effect with degrees of freedom equal to the number of moderator levels minus one; p ! probability; k ! number of samples;
ICC ! mean weighted effect size represented by the intraclass correlation coefficient; T ! tau, estimated standard deviation of the true effect sizes.
were included. Finally, a mixed-effects subgroup analysis comparing published and unpublished studies (albeit limited by the small
number of unpublished studies) indicated that the mean effect was
marginally larger for published studies (g ! 1.59, T ! 0.56, k !
37) than unpublished studies (g ! 0.86, T ! 0.00, k ! 2), QB !
2.91, p ! .09. Thus, the impact of publication bias was not trivial;
nevertheless, the finding that leader stereotypes were more agentic
than communal remains valid.
Analyses for each moderator. Table 7 presents the subgroup
analyses, and Table 8 presents the simple meta-regression models
(left columns). Moderation by participant nationality was not
tested because there were only two non-Western samples. Among
the a priori moderators, publication year (ranging from 1979 to
2007) was significant, with earlier years associated with larger
agency communion differences. Greater percentages of male participants were associated with larger differences, although dividing
the sample by participant sex (leaving only 25 of 47 reports) did
not yield a significant model. Leader domains that were managerial or political were associated with larger differences than the
educational domain, but leader status did not significantly moderate the effect. Percentage of female managers also did not significantly moderate the difference. Among the exploratory moderators, the originators (i.e., Powell or Butterfield) research group
(vs. others) was associated with larger differences, as was the
originators stereotype measure (vs. others).
Multiple meta-regression analysis. Table 8 (right columns)
shows a model that regressed the agency communion differences
onto the variables that were significant predictors in the subgroup
or simple meta-regression models. In this model, the agency
communion difference was larger for managerial and political (vs.
educational) domains and with the originators stereotype measure
and marginally larger in earlier years.
MasculinityFemininity Paradigm
Table 5
Simple Meta-Regressions for the Think ManagerThink
Male Paradigm
Predictor
Year of publication
% Male participants
% Female managers
Participant age
% Male authors
'
49
51
0.01
0.00
0.22
0.04
.09
.77
49
51
"0.00
0.00
"0.22
0.06
.09
.66
48
50
0.00
"0.00
0.18
"0.16
.18
.27
49
51
0.01
"0.00
0.28
"0.04
.03
.75
49
51
"0.00
0.00
"0.29
0.08
.02
.56
Note. Models are random-effects weighted linear regressions calculated

with weights equal to the reciprocal of the variance for each effect size plus
a random-effects component. k ! number of samples; p ! probability; b !
unstandardized regression coefficient; ' ! standardized regression coefficient.
Mean and distribution of effect sizes. Table 3 lists the

seven studies with their characteristics and 101 effect sizes.
Larger gs indicate a more masculine stereotype in relation to the
0.00 that denotes the androgynous midpoint of the scale. If we
regard all of the occupations within each study as separate
samples, nine samples were outliers. Above the mean (i.e.,
masculine direction), the outlier occupations were company
president, U.S. Supreme Court justice, university president,
boat captain, park manager, and orchestra conductor. Below
the mean (i.e., feminine direction), the outlier occupations were
three instances of head librarian. All of these effect sizes were
retained in the analyses because their large distances from the
mean were due to the understandable tendency for very maledominated or female-dominated managerial occupations to
yield extreme means and small standard deviations.
The weighted mean effect of the seven studies, produced by first
averaging across the ratings of different leadership roles within
each study was g ! 0.94 (T ! 0.51; 95% CI [0.56, 1.32]; 90% PI
["0.15, 2.03]). Thus, the overall effect is strongly in the masculine
direction, with participants rating occupations entailing leadership
as quite masculine. There was a large amount of variability in the
effect sizes, Q(6) ! 223.09, p & .001. Nevertheless, I2 was 97.31,
633
Table 6
Multiple Meta-Regressions for the Think ManagerThink Male Paradigm
Variable and class
'
'
Year of publication
% Male participants
Leader status
Participant nationality
Participant age
% Male authors
0.01
"0.00
"0.11
"0.11
0.01
"0.00
0.47
"0.27
"0.10
"0.13
0.47
"0.13
&.001
.03
.37
.27
.001
.31
"0.00
0.00
0.49
0.17
0.00
"0.00
"0.05
0.07
0.43
0.19
0.02
"0.02
.75
.62
.002
.18
.90
.91
Note. Models are random-effects weighted linear regressions calculated with weights equal to the reciprocal of
the variance for each effect size plus a random-effects component. b ! unstandardized regression coefficient;
' ! standardized regression coefficient; p ! probability. k ! 48 for womenleaders similarity and k ! 50 for
menleaders similarity. Codes were as follows: leader status: 1 ! high, 0 ! moderate; participant nationality:
1 ! Eastern, 0 ! Western.
showing that a high proportion of the variability reflected real

differences, not random error.7
Publication bias. It is unclear how publication bias would
operate for this paradigm because the participants rated a heterogeneous set of occupations and the researchers did not compare
managerial occupations to other types of occupations. Thus, there
would be no bias against publication depending on whether managerial occupations were rated as more or less masculine. Nonetheless, we analyzed the distribution of effect sizes for potential
biases in publication or study retrieval. With the funnel plot of the
effect sizes plotted by the standard error with each occupation
represented separately, Eggers test was significant, but the trimand-fill procedure did not suggest that any effect sizes be trimmed.
When we first averaged the occupations within each study, Eggers
test was not significant, and the trim-and-fill method did not
suggest that any studies be trimmed. A cumulative meta-analysis
Table 7
Subgroup Analyses for the AgencyCommunion Paradigm
Variable and class
QB
Participant sex
Male
Female
Leader domain
Educational
Political
Managerial
Leader status
Moderate
High
Research group
Originator
Other
Stereotype measure
Originators
Other
0.00
.99
7.69
0.43
10.13
7.48
25
25
1.20
1.20
0.33
0.52
6
15
26
0.75b
1.48a
1.51a
0.33
0.74
0.54
39
8
1.43
1.27
0.57
0.60
18
29
1.68
1.17
0.64
0.37
37
10
1.55
0.95
0.64
0.48
.02
.51
.001
.006
Note. QB ! between-classes effect with degrees of freedom equal to the

number of moderator levels minus one; p ! probability; k ! number of
samples; d ! mean weighted effect size; T ! tau, estimated standard
deviation of the true effect sizes. Leader domain ds not sharing subscripts
are significantly different from each other at p & .05 when entered into a
mixed-effects model.
from large to small studies indicated very little change in the

overall effect when the smallest study was included. Published and
unpublished studies could not be compared because all studies
were published. Overall, there is thus little evidence of publication
bias.
Analyses for each moderator. Table 9 presents the subgroup
analyses, and Table 10 presents the simple meta-regression models
(left columns). Limited sample variability on many study characteristics precluded testing all but a few as potential moderators. For
example, there were no non-Western samples, all but one sample
had young (i.e., student) participants, and all studies used the same
dependent measure.
Among the a priori moderators, publication year (ranging from
1975 to 2005) was significant, with earlier years associated with
stronger masculinity. Greater percentages of male participants
were also associated with stronger masculinity, although dividing
the samples by participant sex (leaving only 41 of 101 effect sizes)
did not yield a significant model. In terms of leader domain, other
leader roles (boat captain, farm manager, park manager) and
judicial roles were associated with stronger masculinity, followed
by political, arts, and business roles, and finally by educational
roles. High-status (vs. moderate-status) leaders were associated
with stronger masculinity. Smaller percentages of female managers in participants nations were associated with marginally stronger masculinity. Among the exploratory moderators, smaller percentages of male authors were associated with stronger
masculinity.
7
Removing the nine outlying effect sizes still resulted in a significantly
masculine mean weighted effect size of 0.82 (T ! 0.19; 95% CI [0.67,
0.97]; 90% PI [0.41, 1.23]), Q(6) ! 35.44, p & .001; I2 ! 83.07. The
moderator analyses were similar to those reported here, although the
models for leader status and percentage of female managers were nonsignificant. In addition, we computed the effect sizes based on a mean metric,
using the M on the 17 scale as the effect size. We estimated the standard
error of each effect from the standard deviation divided by the square root
of the sample size, yielding a weight of one divided by the squared standard
error (see Lipsey & Wilson, 2001). Using this metric, the mean-weighted
effect size was masculine (M ! 4.91; T ! 0.58; 95% CI [4.48, 5.35]; 90%
PI [3.65, 6.17]), Q(6) ! 4647.70, p & .001; I2 ! 99.87. The moderator
analyses were also very similar using this metric.
634
Table 8
Meta-Regressions for the AgencyCommunion Paradigm
Simple meta-regressions
Multiple meta-regression
Predictor
'
'
Year of publication
% Male participants
Leader domain
% Female managers
Research group
Stereotype measure
47
47
47
47
47
47
"0.04
0.01
"0.38
0.27
.004
.05
"0.02
0.01
0.91
"0.26
0.11
0.43
.08
.41
.004
"0.01
"0.09
.47
"0.01
0.54
"0.01
0.32
.97
.03
the variance for each effect size plus a random-effects component. Blank cells indicate data that are not
applicable to that analysis. k ! number of samples; b ! unstandardized regression coefficient; ' ! standardized
regression coefficient; p ! probability. Codings were as follows: leader domain: 1 ! managerial and political
leaders, 0 ! educational leaders; research group: 1 ! originator, 0 ! other; stereotype measure: 1 ! originators,
0 ! other.
Multiple meta-regression analysis.

Table 10 (right columns) shows a multiple meta-regression model that regressed the
masculinityfemininity effect sizes onto the variables that were at
least marginally significant predictors in the subgroup and simple
meta-regression models. In this model, leader occupations were
more masculine in earlier publication years, in judicial or other
domains, and with high-status roles.
Discussion
All three paradigms showed that stereotypes of leaders are
decidedly masculine. Specifically, people viewed leaders as quite
similar to men but not very similar to women, as more agentic than
communal, and as more masculine than feminine. These findings
are minimally invalidated by publication bias. Only the agency
communion findings appeared somewhat susceptible to exaggeration of the masculinity of the leader stereotype, but this weak bias
Table 9
Subgroup Analyses for the MasculinityFemininity Paradigm
Variable and class
QB
Participant sex
Male
Female
Leader domain
Education
Business
Arts
Political
Judicial
Othera
Leader status
Moderate
High
2.53
.11
18.81
8.36
41
41
0.92
0.67
0.73
0.70
16
48
8
8
8
13
0.49a
0.84a
1.07a,b
1.08a,b
1.80b,c
1.93c
1.21
1.02
0.77
0.37
1.87
0.85
89
12
0.92
1.91
0.96
1.97
.002
.004
Note. QB ! between classes effect with degrees of freedom equal to the

number of moderator levels minus one; p ! probability; k ! number of
samples; d ! mean weighted effect size; T ! tau, estimated standard
deviation of the true effect sizes. Leader domain ds not sharing subscripts
are significantly different from each other at p & .05 when entered into a
mixed-effects model.
a
Other occupations were boat captain, farm manager, park manager.
does not compromise the strong overall masculinity of the leader

stereotype.
In addition to this large and highly robust masculinity effect, our
analyses revealed considerable variability in the extent to which
stereotypes of leaders are masculine versus more androgynous,
although the rarity of negative effect sizes in the agency
communion and masculinityfemininity paradigms indicates that
conceptions of leadership hardly ever strayed into feminine territory. This moderation for the most part was consistent with our
predictions. To examine the extent to which the findings supported
this theory, we briefly review and discuss them across the three
paradigms.
A Priori Moderators
Year of publication. Evidence of the increasing androgyny
of the leader stereotype over publication years emerged in all three
paradigms, including in the multiple regression equations that
controlled for other moderator variables (albeit as a marginal effect
in the agency communion multiple meta-regression). In the think
managerthink male paradigm, the increasing similarity of leaders
and women that emerged in our analyses could be due to change
in women or leader stereotypes or both. Because participants rated
only leaders (and not men or women) in the agency communion
and the masculinityfemininity paradigms, the effects of moderators on these effect sizes could be due only to differing beliefs
about leaders. Given that all three paradigms showed increasing
androgyny of leadership over time and that previous research
indicates little change in stereotypes of women toward greater
masculinity (see Lueptow et al., 2001), the most appropriate conclusion appears to be that the popular conception of leadership has
changed over time. In addition, the think managerthink male
paradigm indicated that change over time took the form of increasing similarity between leaders and women without change in the
similarity between leaders and men. Thus, our conclusion is that
leadership now, more than in the past, appears to incorporate more
feminine relational qualities, such as sensitivity, warmth, and
understanding, thus adding them to the masculine dominance and
strength qualities traditionally associated with leadership.
635
Table 10
Meta-Regressions for the MasculinityFemininity Paradigm
Simple meta-regressions
Multiple meta-regression
Predictor
'
'
Year of publication
% Male participants
Leader domain
Leader status
% Female managers
% Male authors
101
101
101
101
101
101
"0.06
0.08
"0.42
0.22
&.001
.01
"0.02
"0.01
"0.16
"0.19
.06
.02
"0.05
"0.01
1.17
1.20
0.00
"0.00
"0.39
"0.02
0.36
0.30
0.00
"0.02
&.001
.85
&.001
&.001
.98
.76
the variance for each effect size plus a random-effects component. Blank cells indicate data that were not
applicable to that analysis. k ! number of samples; b ! unstandardized regression coefficient; ' ! standardized
regression coefficient; p ! probability. Codings were as follows: leader domain: 1 ! judicial and other, 0 !
education, business, arts, and political; leader status: 1 ! high, 0 ! moderate.
Consistent with existing research (Beaman et al., 2009; Dasgupta & Asgari, 2004), this change may reflect peoples increasing
exposure to women among organizational managers and political
leaders. Yet this change may also reflect the increasing failure of
traditionally masculine, command-and-control styles of leadership
to meet the complex challenges involved in contemporary management and political leadership (Kanter, 1997; Lipman-Blumen,
2000). If so, men leaders may be under pressure to enlarge their
behavioral repertoire to include a greater measure of culturally
feminine relational skills (Eagly & Carli, 2007). Men appear to be
succeeding, given that they are not perceived as any less similar to
leaders than they were in the past, according to our analysis that
separated the role incongruity of men and leaders from that of
women and leaders in the think managerthink male paradigm.
These findings of increased femininity of leadership without decreased masculinity suggest that women leaders would be well
advised to retain elements of a masculine leadership style to avoid
a mismatch with leader roles, even if they now have greater
flexibility to incorporate elements of a feminine leadership style.
Participant sex.
An additional prediction was that men
would have a more masculine construal of leadership than women.
Only the think managerthink male paradigm allowed all of the
findings to be separated or identified by participant sex, and the
resulting analyses yielded smaller womenleaders correlations for
male than female participants. Although this finding was not
replicated in the other paradigms, a related analysis that did not
rely on separating female and male participant samples showed
that a greater proportion of male participants yielded marginally
smaller womenleaders correlations in the think managerthink
male paradigm as well as significantly larger masculine leader
stereotypes in both the agency communion and the masculinity
femininity paradigms. In sum, evidence that men construe leadership as more masculine than women is present in all three paradigms and strongest for mens stereotyping of leaders as not very
similar to women.
This finding is consistent with meta-analytic research showing
that men, but not women, devalued womens leadership in experimental studies that held constant all leader characteristics other
than their sex (see Eagly et al., 1992). Also, mens failure to accord
women many of the qualities of leaders is consistent with their
greater social dominance (e.g., Pratto, Sidanius, & Levin, 2006)
and sexism (e.g., Glick & Fiske, 1996). The implications of mens
reservations about womens leadership abilities are important.
Because men are the more typical holders of decision-making
power in organizations, their beliefs that women lack the qualities
of leaders can limit womens access to positions of authority and
slow their progress into leadership positions.
Leader status. Consistent with the rarity of women in top
positions, higher status leadership positions were expected to have
a more masculine stereotype. Both the think managerthink male
and the masculinityfemininity paradigms found greater masculinity of high-status leadership positions (e.g., upper-level managers) than moderate-status positions (e.g., middle-level managers or
managers in general). The think managerthink male studies suggested that this effect appears mainly because higher status leaders
are stereotypically more similar to men rather than less similar to
women. Although these effects remained significant in the multiple meta-regressions, they should be interpreted cautiously because the studies offered relatively few instances of especially
high-status leader roles. In addition, the effects may be constrained
to certain types of high-status roles, as the agency communion
paradigm, which contained a greater number of political highstatus leadership positions than the other two paradigms, did not
yield this moderation.
According to role congruity theory, women entering higher
status positions should encounter more prejudice because of the
greater stereotypical mismatch between women and leadership.
Our think managerthink male findings suggest instead that men
gain greater congruity as status increases, thereby contributing to
their increasing numerical dominance at higher levels. Yet women
who do manage to be successful in very high-status roles may be
perceived as highly competent, because people assume a double
standard whereby such women had to overcome especially difficult challenges (Rosette & Tost, 2010).
Other a priori variables. Our ability to test the remaining a
priori moderatorsleader domain, nationality of participants, and
percentage of female managerswas compromised by the limitations of the available studies. Because the three paradigms used
different types of leader groups, the effects of leader domain were
not comparable across the paradigms, and the think managerthink
male paradigm did not present sufficient variability in domain to
allow a test of this variable. In the agency communion and
636
masculinityfemininity paradigms, educational positions were the

least masculine, consistent with the female domination of the
education administrator job category (63% women; U.S. Department of Labor, 2010b). In the masculinityfemininity paradigm,
the judicial domain and the category of other leaders were the
most masculine. Judicial occupations (supreme court justice and
federal judge) are highly male dominated: Only four women have
ever served as Supreme Court Justices, and only 9% of federal
judges serving since 1789 have been women, including 30% of
currently active sitting judges (U.S. Courts, 2011). However, the
other leader domain, consisting of boat captain, farm manager,
and park manager, deserves comment. These heterogeneous roles
are likely also male dominated, as men constitute 82% of farm,
ranch, and other agricultural managers and 79% of people in
farming, fishing, and forestry occupations in general (U.S. Department of Labor, 2010b). In addition, these are outdoor occupations
that often require physical strength as well as leadership abilities,
which likely increased their perceived masculinity (Cejka & Eagly,
1999). Future research should systematically investigate stereotypes in leader domains that vary in the representation of women
and other attributes.
Comparisons of effects across nations were limited by the
overrepresentation of the United States and the rarity of studies
from non-Western nations. Nevertheless, consistent with national
data on the status of women (e.g., Hausmann et al., 2009) and on
attitudes toward women as leaders (P. Norris & Inglehart, 2001),
participants from Eastern nations showed a marginally weaker
womenleaders similarity than participants from Western nations
(although this effect was nonsignificant in the multiple regression).
The percentage of female managers in participants nations was a
more informative variable, which was associated with marginally
weaker masculinity in the masculinityfemininity paradigm (although a nonsignificant predictor in the multiple meta-regression).
Yet this variable was not significantly related to masculinity in the
think managerthink male and agency communion paradigms. A
more systematic global representation of nations would provide a
better test of the impact of women in management on leader
stereotypes.
Exploratory Moderators
The exploratory variables yielded a few findings of interest. In
the think managerthink male paradigm, older participants, who
were a mix of managerial and nonmanagerial employees, showed
a stronger womenleaders similarity than younger participants,
who were business or other undergraduate students. Although
these results support the conclusion that students and others with
little workplace experience hold more masculine stereotypes of
leaders, the nonreplication in the agency communion paradigm
renders these findings ambiguous.
The originators of the agency communion paradigm produced
more masculine stereotypes, although this effect was no longer
significant in the multiple regression and not replicated in the other
paradigms. The originators measure may have produced greater
masculinity in the agency communion paradigm, because this
measure, the Bem Sex Role Inventory, had better psychometric
properties than many of the other measures, which were often
more casually constructed without a factor or item analyses. Yet
the rarity of reporting measures internal consistencies in the
primary studies precluded a statistical test of the idea that the Bem
Sex Role Inventory provided more coherent measures of agency
and communion. The percentage of male authors had inconsistent
effects across the paradigms and was not a significant predictor the
multiple meta-regressions. In general, the lack of consistent moderation of these paradigms by variables related to the authors of the
primary studies and to stereotype measures used in the studies
indicates that the effects revealed in this meta-analysis are not
limited to certain research groups or dependent measures.
Strengths and Limitations of the Paradigms

Our three-part meta-analysis demonstrates the advantages of
examining leader stereotypes from the perspective of differing
research paradigms. The fact that the paradigms conceptually
replicate one another lends confidence to results that are similar
across the paradigms. However, each paradigm has its own
strengths and limitations.
The think managerthink male paradigm has the advantage of
comparing the cultural construal of leadership to that of men and
women separately, thus allowing for leadership to be seen as
similar to men, women, both, or neither. The ICC also provides a
direct measure of the similarity between female and male stereotypes and leader stereotypes, yielding a clear-cut test of role
congruity theory (Eagly & Karau, 2002). The downside of this
method is that changes in intraclass correlations associated with
moderators such as publication year do not indicate whether gender or leader stereotypes (or both) have changed.
The agency communion paradigm, on the other hand, assesses only leader stereotypes. The scales used in the paradigm
consist of agentic and communal traits and thus assess the
perceived masculinity of leadership relative to its perceived
femininity, given that agency is stereotypically masculine and
communion is stereotypically feminine (Wood & Eagly, 2010).
Any changes in the content of gender stereotypes are not taken
into account in this assessment.
The masculinityfemininity paradigm is the most limited of the
paradigms because its bipolar measure does not allow for leader
stereotypes to be both masculine and feminine. In addition, the fact
that the effect sizes within most samples were dependent is a
limitation in relation to the moderator analyses, which treated these
effect sizes as independent to take account of the wide variation in
the types of leader roles but lessened the accuracy of these analyses. However, because participants rated a wide variety of occupations in most samples, only some of which involved leadership,
the method mirrors natural settings in which people observe many
different types of occupational roles.
A consideration for all three paradigms is the extent to which
participants responses might have been contaminated by selfpresentational pressures or studies demand characteristics.
Fortunately, the think managerthink male paradigm is relatively subtle, because with very few exceptions, participants
rated only women, men, or leaders in a between-subjects design. Participants could not have discerned that the study had to
do with the correspondence of gender and leader stereotypes. In
contrast, in the agency communion studies, participants rated
leaders on masculine and feminine attributes, and in the
masculinityfemininity studies, participants rated leaders on an
explicit masculinityfemininity scale. These more obvious de-
signs might foster self-presentational pressures to appear tolerant of female leaders by downplaying the masculinity of leader
roles. Yet such pressures would have been mitigated in the
masculinityfemininity studies by the presentation of a wide
range of occupations within which only relatively few pertained
to leadership. Nevertheless, future research should include
more subtle measures that assess implicit associations between
gender and leadership. In one study that included implicit
measures, Rudman and Kilanski (2000) found that participants,
especially men, more quickly paired male names with authority
roles and female names with subordinate roles and were quicker
to respond to positive words after seeing a male than a female
authority figure (see also Beaman et al., 2009; Dasgupta &
Asgari, 2004). Although these findings are consistent with the
masculine construal of leadership found in our meta-analysis,
additional research incorporating implicit measures would be
informative.
The current meta-analysis also highlights other areas for
future research that are not well addressed by the current
primary studies. For example, more studies in Eastern nations
would help to address questions about the masculinity of leadership roles across cultures. Also, although research has examined role incongruity based on race (Sy et al., 2010), it has not
addressed intersectionality, whereby the role incongruity of
women in relation to leadership may depend on other group
memberships (e.g., race, ethnicity, sexual orientation) as well as
their gender (Cole, 2009). In addition, expanding the domain of
research beyond managerial leadership positions would be useful in understanding the contextual nature of stereotypes of
leadership. More primary research on the impact of leader
status is also needed, and greater variation of research designs
and measures would be appropriate. Finally, additional metaanalyses should assess future changes and contextual variation
in the cultural construal of leadership.
Implications of the Masculinity of Leadership

This meta-analysis establishes that the masculinity of the cultural stereotype of leadership is a large effect that is robust across
variation in many aspects of leaders social contexts. According to
role congruity theory (Eagly & Karau, 2002), this representation of
leadership poses a problem for women because female stereotypes
do not match expectations for leaders. Even women who possess
outstanding qualifications for leadership may have the burden of
overcoming preconceptions that they are not well equipped to lead.
Not only do the descriptive aspects of stereotyping make it difficult for women to gain access to leader roles, but the prescriptive
aspects of stereotyping could produce conflicting expectations
concerning how female leaders should behavethat is, that they
should be agentic to fulfill the leader role but communal to fulfill
the female gender role. Thus, although women leaders may be seen
as competent, women who disregard their communal gender role
are often disliked and therefore still the recipients of prejudice
even though they fulfill their leader role (Rudman & Glick, 1999,
2001).
These cross-pressures are likely to produce a double bind that
discourages women from presenting themselves in ways that
others consider too masculine or too feminine (Eagly & Carli,
2007), constraining their behavior to an androgynous middle
637
ground. The tendencies for women leaders, compared with men

leaders, to manifest a somewhat more transformational leadership style and to wield rewards as incentives (Eagly,
Johannesen-Schmidt, & van Engen, 2003), which are not distinctively masculine or feminine behaviors, may reflect one
way for women to satisfy both their leader role and their gender
role. Because some of the elements of transformational leadership, especially the mentoring and empowering of subordinates,
appear to be aligned more with the feminine than the masculine
gender role, findings suggest that transformational leadership is
in general androgynous or even slightly feminine (Duehr &
Bono, 2006; Hackman, Furniss, Hills, & Paterson, 1992). Given
the demonstrated effectiveness of these aspects of leadership
style (Judge & Piccolo, 2004), the relatively androgynous behavioral repertoire that is common in women leaders should
ironically facilitate their success in leader roles. Thus, consistent with this meta-analysis, womens inclination toward an
androgynous leadership style should often help them fulfill both
their gender role and their leader role, given that the conception
of leadership has incorporated more feminine qualities in more
recent years, in moderate-status leader roles, in certain occupational domains, and in the minds of women.
Although role congruity theory describes how the masculinity of leadership influences prejudice toward women leaders,
other research suggests that these stereotypes may also affect
women themselves, sometimes decreasing their performance as
leaders and their identification with leadership. Evidence of
such outcomes has emerged in research on stereotype threat.
Specifically, when women were reminded of general female
stereotypes by watching gender-stereotypical (vs. neutral) commercials, they were less likely to express interest in being a
leader on a subsequent group task unless they were also told
that research had found no sex differences in leadership (Davies, Spencer, & Steele, 2005). Nonetheless, activation of a
masculine leader stereotype can also cause women to react
against the stereotype by showing greater confidence and better
leader performance (Hoyt & Blascovich, 2007; Hoyt, Johnson,
Murphy, & Skinnell, 2010). However, positive responses of this
type tended to change to lower self-appraisals and poorer leadership performance for women who lacked confidence in their
capability as leaders (Hoyt & Blascovich, 2007) or who were
exposed to a double threat from, for example, stereotype activation combined with solo status in an otherwise all-male group
(Hoyt et al., 2010).
In conclusion, this meta-analysis establishes a strong and robust
tendency for leadership to be viewed as culturally masculine
across three paradigms that use different methods. The implications of the masculinity of leader roles for prejudice against female
leaders are straightforward: Men fit cultural construals of leadership better than women do and thus have better access to leader
roles and face fewer challenges in becoming successful in them.
Despite some overall change toward more androgynous beliefs
about leadership, stereotyping continues to contribute to the labyrinthine challenges that women encounter in attaining roles that
yield substantial power and authority. Given the strongly masculine cultural stereotype of leadership quantified by this metaanalysis, these challenges are likely to continue for some time to
come.
638
References
References marked with an asterisk indicate studies included in the
meta-analyses.
Atwater, L. E., Brett, J. F., Waldman, D., DiMare, L., & Hayden, M. V.
(2004). Mens and womens perceptions of the gender typing of management subroles. Sex Roles, 50, 191199. doi:10.1023/B:
SERS.0000015551.78544.35
Avolio, B. J. (1999). Full leadership development: Building the vital forces
in organizations. Thousand Oaks, CA: Sage.
Beaman, L., Chattopadhyay, R., Duflo, E., Pande, R., & Topalova, P.
(2009). Powerful women: Does exposure reduce bias? Quarterly Journal
of Economics, 124, 14971540. doi:10.1162/qjec.2009.124.4.1497
*Beggs, J. M., & Doolittle, D. C. (1993). Perceptions now and then of
occupational sex typing: A replication of Shinars (1975) study. Journal
of Applied Social Psychology, 23, 14351453. doi:10.1111/j.15591816.1993.tb01042.x
Bem, S. L. (1974). The measurement of psychological androgyny. Journal
of Consulting and Clinical Psychology, 42, 155162. doi:10.1037/
h0036215
Bennis, W. (1989). On becoming a leader. Reading, MA: Addison-Wesley.
Bianchi, S. M., Robinson, J. P., & Milkie, M. A. (2006). Changing rhythms
of American family life. New York, NY: Russell Sage.
*Booysen, L., & Nkomo, S. (2006). Think managerthink (fe)male: A
South African perspective. International Journal of the Interdisciplinary
Social Sciences, 1, 2334.
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009).
Introduction to meta-analysis. Chichester, England: Wiley. doi:10.1002/
9780470743386
*Boyce, L. A., & Herd, A. M. (2003). The relationship between gender role
stereotypes and requisite military leadership characteristics. Sex Roles,
49, 365378. doi:10.1023/A:1025164221364
*Brenner, O. C., Tomkiewicz, J., & Schein, V. E. (1989). The relationship
between sex role stereotypes and requisite management characteristics
revisited. Academy of Management Journal, 32, 662 669. doi:10.2307/
256439
Broverman, I. K., Broverman, D. M., Clarkson, F. E., Rosenkrantz, P. S.,
& Vogel, S. R. (1970). Sex-role stereotypes and clinical judgments of
mental health. Journal of Consulting and Clinical Psychology, 34, 17.
doi:10.1037/h0028797
Burgess, D., & Borgida, E. (1999). Who women are, who women should
be: Descriptive and prescriptive gender stereotyping in sex discrimination. Psychology, Public Policy, and Law, 5, 665 692. doi:10.1037/
1076-8971.5.3.665
*Butterfield, D. A., & Powell, G. N. (1981). Convergent validity in
students perceptions of Jimmy Carter, Ted Kennedy, and the ideal
president. Perceptual and Motor Skills, 52, 5156.
Buttner, E. H., & Rosen, B. (1988). Bank loan officers perceptions of the
characteristics of men, women, and successful entrepreneurs. Journal of
Business Venturing, 3, 249 258. doi:10.1016/0883-9026(88)90018-3
*Byler, V. L. J. (2000). Perceptions of effective athletic administrators and
gender schemas. Dissertation Abstracts International, 61(12), 4688.
*Cann, A., & Siegfried, W. D. (1987). Sex stereotypes and the leadership
role. Sex Roles, 17, 401 408. doi:10.1007/BF00288143
Catalyst. (2010). Quick takes: Women in U.S. management. Retrieved from
http://www.catalyst.org/publication/206/women-in-us-management
Cejka, M. A., & Eagly, A. H. (1999). Gender-stereotypic images of
occupations correspond to the sex segregation of employment. Personality and Social Psychology Bulletin, 25, 413 423. doi:10.1177/
0146167299025004002
Center for American Women and Politics. (2010). Fact sheet: History of
women governors. Retrieved from http://www.cawp.rutgers.edu/
fast_facts/levels_of_office/documents/govhistory.pdf
Center for American Women and Politics. (2011). Fact sheet: Women in
the U.S. Congress 2011. Retrieved from http://www.cawp.rutgers.edu/

fast_facts/levels_of_office/documents/cong.pdf
Cialdini, R. B., & Trost, M. R. (1998). Social influence: Social norms,
conformity, and compliance. In D. Gilbert, S. Fiske, & G. Lindzey
(Eds.), The handbook of social psychology (Vol. 2; 4th ed., pp. 151
192). New York, NY: McGraw-Hill.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed.). Hillsdale, NJ: Erlbaum.
*Cohen-Kaner, I. (1993). Attributions about male and female leaders in
organizations. Dissertation Abstracts International, 53(11-B), 6021.
Cole, E. R. (2009). Intersectionality and research in psychology. American
Psychologist, 64, 170 180. doi:10.1037/a0014564
Comprehensive Meta-Analysis (Version 2.2.050) [Computer software].
Englewood, NJ: Biostat.
Constantinople, A. (1973). Masculinityfemininity: An exception to a
famous dictum? Psychological Bulletin, 80, 389 407. doi:10.1037/
h0035334
Couch, J. V., & Sigler, J. N. (2001). Gender perceptions of professional
occupations. Psychological Reports, 88, 693 698. doi:10.2466/
PR0.88.3.693-698
Dasgupta, N., & Asgari, S. (2004). Seeing is believing: Exposure to
counterstereotypic women leaders and its effect on the malleability of
automatic gender stereotyping. Journal of Experimental Social Psychology, 40, 642 658. doi:10.1016/j.jesp.2004.02.003
Davies, P. G., Spencer, S. J., & Steele, C. M. (2005). Clearing the air:
Identity safety moderates the effects of stereotype threat on womens
leadership aspirations. Journal of Personality and Social Psychology,
88, 276 287. doi:10.1037/0022-3514.88.2.276
*Dion, K. L., & Schuller, R. A. (1990). Ms. and the manager: A tale of two
stereotypes. Sex Roles, 22, 569 577. doi:10.1007/BF00288235
*Dion, K. L., & Schuller, R. A. (1991). The Ms. stereotype: Its generality
and its relation to managerial and marital status stereotypes. Canadian
Journal of Behavioural Science, 23, 25 40. doi:10.1037/h0078962
*Dodge, K. A., Gilroy, F. D., & Fenzel, L. M. (1995). Requisite management characteristics revisited: Two decades later. Journal of Social
Behavior & Personality, 10, 253264.
DSTAT: Software for the meta-analytic review of research literatures
(Version 1.11) [Computer software]. Hillsdale, NJ: Erlbaum.
*Duehr, E. E., & Bono, J. E. (2003). [Stereotypes of managers in a
non-manager sample]. Unpublished raw data.
*Duehr, E. E., & Bono, J. E. (2006). Men, women, and managers: Are
stereotypes finally changing? Personnel Psychology, 59, 815 846. doi:
10.1111/j.1744-6570.2006.00055.x
Eagly, A. H., & Carli, L. L. (1981). Sex of researchers and sex-typed
communications as determinants of sex differences in influenceability:
A meta-analysis of social influence studies. Psychological Bulletin, 90,
120. doi:10.1037/0033-2909.90.1.1
Eagly, A. H., & Carli, L. L. (2007). Through the labyrinth: The truth about
how women become leaders. Boston, MA: Harvard Business School
Press.
Eagly, A. H., & Diekman, A. B. (2005). What is the problem? Prejudice as
an attitude-in-context. In J. F. Dovidio, P. Glick, & L. Rudman (Eds.),
On the nature of prejudice: Fifty years after Allport (pp. 19 35).
Malden, MA: Blackwell.
Eagly, A. H., Johannesen-Schmidt, M. C., & van Engen, M. L. (2003).
Transformational, transactional, and laissez-faire leadership styles: A
meta-analysis comparing women and men. Psychological Bulletin, 129,
569 591. doi:10.1037/0033-2909.129.4.569
Eagly, A. H., & Karau, S. J. (1991). Gender and the emergence of leaders:
A meta-analysis. Journal of Personality and Social Psychology, 60,
685710. doi:10.1037/0022-3514.60.5.685
Eagly, A. H., & Karau, S. J. (2002). Role congruity theory of prejudice
toward female leaders. Psychological Review, 109, 573598. doi:
10.1037/0033-295X.109.3.573

Eagly, A. H., Makhijani, M. G., & Klonsky, B. G. (1992). Gender and the
evaluation of leaders: A meta-analysis. Psychological Bulletin, 111,
322. doi:10.1037/0033-2909.111.1.3
Eagly, A. H., & Mladinic, A. (1994). Are people prejudiced against
women? Some answers from research on attitudes, gender stereotypes,
and judgment of competence. In W. Stroebe & M. Hewstone (Eds.),
European review of social psychology (Vol. 5, pp. 135). New York,
NY: Wiley.
Eagly, A. H., & Sczesny, S. (2009). Stereotypes about women, men, and
leaders: Have times changed? In M. Barreto, M. K. Ryan, & M. T.
Schmitt (Eds.), The glass ceiling in the 21st century: Understanding
barriers to gender equality (pp. 21 47). Washington, DC: American
Psychological Association. doi:10.1037/11863-002
Epitropaki, O., & Martin, R. (2004). Implicit leadership theories in applied
settings: Factor structure, generalizability, and stability over time. Journal of Applied Psychology, 89, 293310. doi:10.1037/00219010.89.2.293
European Commission. (2010). Gender balance in decision making. Retrieved from http://ec.europa.eu/social/main.jsp?catId!764&langId!en
*Fernandes, E., & Cabral-Cardoso, C. (2003). Gender asymmetries and the
manager stereotype among management students. Women in Management Review, 18, 77 87. doi:10.1108/09649420310462352
*Foster, F. (1994). Managerial sex role stereotyping among academic staff
within UK business schools. Women in Management Review, 9, 1722.
doi:10.1108/09649429410056308
*Fullagar, C. J., Sverke, M., Sumer, H. C., & Slick, R. (2003). Managerial sex-role
stereotyping: A cross cultural analysis. International Journal of Cross Cultural
Management, 3, 93107. doi:10.1177/1470595803003001112
Gallup News Service. (2001, February 21). Americans see women as
emotional and affectionate, men as more aggressive: Gender specific
stereotypes persist in recent Gallup poll [News release]. Retrieved from
http://www.gallup.com/poll/1978/americans-see-women-emotionalaffectionate-men-more-aggressive.aspx
*Gatton, D. S., DuBois, C. L. Z., & Faley, R. H. (1999). The effects of
organizational context on occupational gender-stereotyping. Sex Roles,
40, 567582. doi:10.1023/A:1018888013808
Gergen, D. (2005). Women leading in the twenty-first century. In L.
Coughlin, E. Wingard, & K. Hollihan (Eds.), Enlightened power: How
women are transforming the practice of leadership (pp. xv-xxix). San
Francisco, CA: Jossey-Bass.
Gerstner, C. R., & Day, D. V. (1994). Cross-cultural comparison of leader
prototypes. The Leadership Quarterly, 5, 121134. doi:10.1016/10489843(94)90024-8
Glick, P. (1991). Trait-based and sex-based discrimination in occupational
prestige, occupational salary, and hiring. Sex Roles, 25, 351378. doi:
10.1007/BF00289761
Glick, P., & Fiske, S. T. (1996). The Ambivalent Sexism Inventory:
Differentiating hostile and benevolent sexism. Journal of Personality
and Social Psychology, 70, 491512. doi:10.1037/0022-3514.70.3.491
Glick, P., & Fiske, S. T. (2007). Sex discrimination: The psychological
approach. In F. J. Crosby, M. S. Stockdale, & S. A. Ropp (Eds.), Sex
discrimination in the workplace: Multidisciplinary perspectives (pp.
155187). Malden, MA: Blackwell.
Gutek, B. A., & Stevens, D. A. (1979). Effects of sex of subject, sex of
stimulus cue, and androgyny level on evaluations in work situations
which evoke sex role stereotypes. Journal of Vocational Behavior, 14,
2332. doi:10.1016/0001-8791(79)90047-2
Hackman, M. Z., Furniss, A. H., Hills, M. J., & Paterson, T. J. (1992).
Perceptions of gender-role characteristics and transformational and
transactional leadership behaviours. Perceptual and Motor Skills, 75,
311319. doi:10.2466/PMS.75.4.311-319
Hausmann, R., Tyson, L. D., & Zahidi, S. (2009). The global gender gap
report 2009. Geneva, Switzerland: World Economic Forum.
639
Heilman, M. E. (1983). Sex bias in work settings: The lack of fit model.
Research in Organizational Behavior, 5, 269 298.
Heilman, M. E. (2001). Description and prescription: How gender stereotypes prevent womens ascent up the organizational ladder. Journal of
Social Issues, 57, 657 674. doi:1.1111/0022-4537.00234
*Heilman, M. E., Block, C. K., Martell, R. F., & Simon, M. C. (1989). Has
anything changed? Current characteristics of men, women, and managers. Journal of Applied Psychology, 74, 935942. doi:10.1037/00219010.74.6.935
Heilman, M. E., & Eagly, A. H. (2008). Gender stereotypes are alive, well, and
busy producing workplace discrimination. Industrial and Organizational
Psychology, 1, 393398. doi:10.1111/j.1754-9434.2008.00072.x
Heilman, M. E., & Parks-Stamm, E. J. (2007). Gender stereotypes in the
workplace: Obstacles to womens career progress. In S. J. Correll (Ed.),
Social psychology of gender: Advances in group processes (Vol. 24, pp.
4777). Oxford, England: Elsevier.
Hogue, M., & Lord, R. G. (2007). A multilevel, complexity theory approach to understanding gender bias in leadership. The Leadership
Quarterly, 18, 370 390. doi:10.1016/j.leaqua.2007.04.006
Howell, S. E., & Day, C. L. (2000). Complexities of the gender gap.
Journal of Politics, 62, 848 874. doi:10.1111/0022-3816.00036
Hoyt, C., & Blascovich, J. (2007). Leadership efficacy and women leaders
responses to stereotype activation. Group Processes & Intergroup Relations, 10, 595 616. doi:10.1177/1368430207084718
Hoyt, C. L., Johnson, S. K., Murphy, S. E., & Skinnell, K. H. (2010). The
impact of blatant stereotype activation and group sex-composition on
female leaders. The Leadership Quarterly, 21, 716 732. doi:10.1016/
j.leaqua.2010.07.003
*Huddy, L., & Terkildsen, N. (1993). The consequences of gender stereotypes for women candidates at different levels and types of organizations. Political Research Quarterly, 46, 503525. doi:10.1177/
106591299304600304
*Inderlied, S. D., & Powell, G. (1979). Sex-role identity and leadership
style: Different labels for the same concept? Sex Roles, 5, 613 625.
doi:10.1007/BF00287664
International Labour Organization. (1985). Yearbook of labour statistics.
Geneva, Switzerland: Author.
Inter-Parliamentary Union. (2010). Women in national parliaments. Retrieved from http://www.ipu.org/wmn-e/classif.htm
Johnson, B. T., & Eagly, A. H. (1989). Effects of involvement on persuasion: A meta-analysis. Psychological Bulletin, 106, 290 314. doi:
10.1037/0033-2909.106.2.290
Johnson, S. K., Murphy, S. E., Zewdie, S., & Reichard, R. J. (2008). The
strong, sensitive type: Effects of gender stereotypes and leadership
prototypes on the evaluation of male and female leaders. Organizational
Behavior and Human Decision Processes, 106, 39 60. doi:10.1016/
j.obhdp.2007.12.002
Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional
leadership: A meta-analytic test of their relative validity. Journal of
Applied Psychology, 89, 755768. doi:10.1037/0021-9010.89.5.755
Kanter, R. M. (1997). Rosabeth Moss Kanter on the frontiers of management. Boston, MA: Harvard Business School Press.
*Karau, S. J. (2005). [Gender role stereotypes and the requisite characteristics of middle managers, CEOs, and entrepreneurs]. Unpublished raw
data.
*Karau, S. J., & Elsaid, A. M. (2005). [Cultural influences on perceptions
of female managers: A comparison of Egypt and the United States].
Unpublished raw data.
*Karau, S. J., & Hansen, E. (2005). [Cultural influences on perceptions of
female managers: A comparison of Sweden and the United States].
Unpublished raw data.
Kelley, L. R., & Blashfield, R. K. (2009). An example of psychological
sciences failure to self-correct. Review of General Psychology, 13,
122129. doi:10.1037/a0015287
640
*Kent, J. B. (1984). Relationships among gender stereotypes and requisite

school administrator characteristics as perceived by principals in the
Commonwealth of Virginia. Dissertation Abstracts International, 46(1A), 3536.
*Koch, S. C., Luft, R., & Kruse, L. (2005). Women and leadership20
years later: A semantic connotation study. Social Science Information,
44, 9 39. doi:10.1177/0539018405050433
*Kunkel, A., Dennis, M. R., & Waters, E. (2003). Contemporary university
students ratings of characteristics of men, women, and CEOs. Psychological Reports, 93, 11971213. doi:10.2466/PR0.93.8.1197-1213
Langford, T., & MacKinnon, N. J. (2000). The affective bases for the
gendering of traits: Comparing the United States and Canada. Social
Psychology Quarterly, 63, 34 48. doi:10.2307/2695879
Leaper, C., Anderson, K. J., & Sanders, P. (1998). Moderators of gender
effects on parents talk to their children: A meta-analysis. Developmental Psychology, 34, 327. doi:10.1037/0012-1649.34.1.3
Lipman-Blumen, J. (2000). Connective leadership: Managing in a changing world. New York, NY: Oxford University Press.
Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand
Oaks, CA: Sage.
*Lord, R. G., Foti, R. J., & de Vader, C. L. (1984). A test of leadership
categorization theory: Internal structure, information processing, and
leadership perceptions. Organizational Behavior and Human Decision
Processes, 34, 343378. doi:10.1016/0030-5073(84)90043-6
Lord, R. G., & Maher, K. J. (1993). Leadership and information processing: Linking perceptions and performance. New York, NY: Routledge.
Lueptow, L. B., Garovich-Szabo, L., & Lueptow, M. B. (2001). Social
change and the persistence of sex typing: 1974 1997. Social Forces, 80,
136. doi:10.1353/sof.2001.0077
*Luft, R. (2003). Ist Fuhrung noch mannlich? Eine sprachpsychologische
Untersuchung mit dem semantischen Differential. [Is leadership still
masculine? A psycholinguistic investigation using the semantic differential]. (Unpublished doctoral dissertation.) Heidelberg University,
Heidelberg, Germany.
Lyness, K. S., & Heilman, M. E. (2006). When fit is fundamental: Performance evaluations and promotions of upper-level female and male
managers. Journal of Applied Psychology, 91, 777785. doi:10.1037/
0021-9010.91.4.777
*Mahoney, T. A., & Blake, R. H. (1987). Judgements of appropriate
occupational pay as influenced by occupational characteristics and sex
characterisations. Applied Psychology: An International Review, 36,
2538. doi:10.1111/j.1464-0597.1987.tb00372.x
*Maier, M. (1993). Revisiting (and resolving?) the androgyny/masculinity
debate in management. The Journal of Mens Studies, 2, 157171.
doi:10.3149/jms.0202.157
*Marmenout, K. (2008). Women-focused leadership development in the
Middle East: Generating local knowledge. Retrieved from Social Science Research Network website, http://ssrn.com/abstract!1427729
*Marongiu, S., & Ekehammar, B. (2000). A gender perspective on person
manager fit and managerial advancement. Scandinavian Journal of
Psychology, 41, 175180. doi:10.1111/1467-9450.00185
*Massengill, D., & Di Marco, N. (1979). Sex-role stereotypes and requisite
management characteristics: A current replication. Sex Roles, 5, 561
570. doi:10.1007/BF00287660
McCauley, C. D. (2004). Successful and unsuccessful leadership. In J.
Antonakis, A. T. Cianciolo, & R. J. Sternberg (Eds.), The nature of
leadership (pp. 199 221). Thousand Oaks, CA: Sage.
McGraw, K. O., & Wong, S. P. (1996). Forming inferences about some
intraclass correlation coefficients. Psychological Methods, 1, 30 46.
doi:10.1037/1082-989X.1.1.30
McTague, T., Stainback, K., & Tomaskovic-Devey, D. (2009). An organizational approach to understanding sex and race segregation in U.S.
workplaces. Social Forces, 87, 1499 1527. doi:10.1353/sof.0.0170
*Mook, K. (2005). Think manager, think male: Onderzoek naar de
(on)bewuste gedachtegang over managers [Think manager, think

male: Investigation of (un)conscious thoughts about managers]. (Unpublished masters thesis). Tilburg University, Tilburg, the Netherlands.
*Munoz Sastre, T. M., Fouguereau, E., Igier, V., Salvatore, N., & Mullet,
E. (2000). Perception of occupational gender typing: A replication on
European samples of Shinars (1975) and Beggs and Doolittles (1993)
studies. Journal of Applied Social Psychology, 30, 430 441. doi:
10.1111/j.1559-1816.2000.tb02324.x
*Neergaard, H., Madsen, H., Hersby, M., Ulhi, J. P., & Toft Madsen, M.
(2007). Revisiting the think managerthink male hypothesis in an egalitarian culture. Manuscript submitted for publication.
*Norris, J. M., & Wylie, A. M. (1995). Gender stereotyping of the managerial
role among students in Canada and the United States. Group & Organization Management, 20, 167182. doi:10.1177/1059601195202006
Norris, P., & Inglehart, R. (2001). Women and democracy: Cultural obstacles to equal representation. Journal of Democracy, 12, 126 140.
doi:10.1353/jod.2001.0054
Offermann, L., Kennedy, J. K., & Wirtz, P. W. (1994). Implicit leadership
theories: Content, structure, and generalizability. The Leadership Quarterly, 5, 4358. doi:10.1016/1048-9843(94)90005-1
*Orser, B. (1994). Sex role stereotypes and requisite management characteristics: An international perspective. Women in Management Review,
9, 1119. doi:10.1108/09649429410062211
*Pardine, P., Fox, I., & Salzano, J. (1995). Substereotypes and requisite
management characteristics. Retrieved from ERIC database.
(ED397239)
Petty, M. M., & Miles, R. H. (1976). Leader sex-role stereotyping in a
female-dominated work culture. Personnel Psychology, 29, 393 404.
doi:10.1111/j.1744-6570.1976.tb00423.x
*Powell, G. N. (1992). The good manager: Business students stereotypes
of Japanese managers versus stereotypes of American managers. Group
& Organization Management, 17, 44 56. doi:10.1177/
1059601192171004
*Powell, G. N., & Butterfield, D. A. (1979). The good manager: Masculine or androgynous? Academy of Management Journal, 22, 395 403.
doi:10.2307/255597
*Powell, G. N., & Butterfield, D. A. (1984). If good managers are
masculine, what are bad managers? Sex Roles, 10, 477 484. doi:
10.1007/BF00287256
*Powell, G. N., & Butterfield, D. A. (1987). Is the presidential image
reserved for males? Sex-role stereotypes and the 1984 presidential
election. Psychological Reports, 61, 491 495.
*Powell, G. N., & Butterfield, D. A. (1989). The good manager: Did
androgyny fare better in the 1980s? Group & Organization Management, 14, 216 233. doi:10.1177/105960118901400209
*Powell, G. N., Butterfield, D. A., & Parent, J. D. (2002). Gender and
managerial stereotypes: Have the times changed? Journal of Management, 28, 177193. doi:10.1177/014920630202800203
*Powell, G. N., & Kido, Y. (1994). Managerial stereotypes in a global
economy: A comparative study of Japanese and American students
perspectives. Psychological Reports, 74, 219 226.
Pratto, F., Sidanius, J., & Levin, S. (2006). Social dominance theory and
the dynamics of intergroup relations: Taking stock and looking forward.
In W. Stroebe & M. Hewstone (Eds.), European Review of Social
Psychology (Vol. 17, pp. 271320). New York, NY: Psychology Press.
*Rosenwasser, S. M., & Dean, N. G. (1989). Gender role and political
office: Effects of perceived masculinity/femininity of candidate and
political office. Psychology of Women Quarterly, 13, 77 85. doi:
10.1111/j.1471-6402.1989.tb00986.x
Rosette, A. S., & Tost, L. P. (2010). Agentic women and communal
leadership: How role prescriptions confer advantage to top women
leaders. Journal of Applied Psychology, 95, 221235. doi:10.1037/
a0018204
Rudman, L. A., & Glick, P. (1999). Feminized management and backlash

toward agentic women: The hidden costs to women of a kinder, gentler
image of middle managers. Journal of Personality and Social Psychology, 77, 1004 1010. doi:10.1037/0022-3514.77.5.1004
Rudman, L. A., & Glick, P. (2001). Prescriptive gender stereotypes and
backlash toward agentic women. Journal of Social Issues, 57, 743762.
doi:10.1111/0022-4537.00239
Rudman, L. A., & Kilanski, S. E. (2000). Implicit and explicit attitudes
toward female authority. Personality and Social Psychology Bulletin, 26,
13151328. doi:10.1177/0146167200263001
*Rustemeyer, R., & Thrien, S. (1989). Die Managerin der Manager: Wie
weiblich duerfen sie sein, wie maennlich muessen sie sein? [The female
managerthe male manager: How feminine are they allowed to be? How
masculine do they have to be?]. Zeitschrift Fur Arbeits- und Organisationspsychologie, 33, 108 116.
*Ryan, M. K., Haslam, S. A., Hersby, M. D., & Bongiorno, R. (in press).
Think crisisthink female? The glass cliff and contextual variation in the
think managerthink male stereotype. Journal of Applied Psychology.
Sanchez-Meca, J., Marn-Martinez, F., & Chancon-Moscoso, S. (2003).
Effect-size indices for dichotomized outcomes in meta-analysis. Psychological Methods, 8, 448 467. doi:10.1037/1082-989X.8.4.448
*Sauers, D. A., Kennedy, J. C., & OSullivan, D. (2002). Managerial sex
role stereotyping: A New Zealand perspective. Women in Management
Review, 17, 342347. doi:10.1108/09649420210445794
*Schein, V. E. (1973). The relationship between sex role stereotypes and
requisite management characteristics. Journal of Applied Psychology,
57, 95100. doi:10.1037/h0037128
*Schein, V. E. (1975). The relationship between sex role stereotypes and
requisite management characteristics among female managers. Journal
of Applied Psychology, 60, 340 344. doi:10.1037/h0076637
Schein, V. E. (2001). A global look at psychological barriers to womens
progress in management. Journal of Social Issues, 57, 675 688. doi:
10.1111/0022-4537.00235
*Schein, V. E., & Davidson, M. J. (1993). Think managerthink male:
Managerial sex typing among U.K. business students. Management
Development Review, 6(3), 24 28. doi:10.1108/EUM0000000000738
*Schein, V. E., & Mueller, R. (1992). Sex role stereotyping and requisite
management characteristics: A cross-cultural look. Journal of Organizational Behavior, 13, 439 447. doi:10.1002/job.4030130502
*Schein, V. E., Mueller, R., & Jacobson, C. (1989). The relationship
between sex role stereotypes and requisite management characteristics
among college students. Sex Roles, 20, 103110. doi:10.1007/
BF00288030
*Schein, V. E., Mueller, R., Lituchy, T., & Liu, J. (1996). Think manager
think male: A global phenomenon? Journal of Organizational Behavior, 17,
33 41. doi:10.1002/(SICI)1099-1379(199601)17:1&33::AID-JOB778(
3.0.CO;2-F
*Sczesny, S. (2003a). A closer look beneath the surface: Various facets of
the think-managerthink-male stereotype. Sex Roles, 49, 353363. doi:
10.1023/A:1025112204526
*Sczesny, S. (2003b). Fuhrungskompetenz: Selbst- and Fremdwahrnehmung weiblicher und mannlicher Fuhrungskrafte [Leadership competence: Self- and other-perceptions of male and female managers].
Zeitschrift fur Sozialpsychologie, 34, 133145. doi:10.1024//00443514.34.3.133
*Shinar, E. H. (1975). Sexual stereotypes of occupations. Journal of
Vocational Behavior, 7, 99 111. doi:10.1016/0001-8791(75)90037-8
Spence, J. T., & Buckner, C. E. (2000). Instrumental and expressive traits,
trait stereotypes, and sexist attitudes. Psychology of Women Quarterly,
24, 44 62. doi:10.1111/j.1471-6402.2000.tb01021.x
Stainback, K., & Tomaskovic-Devey, D. (2009). Intersections of power
and privilege: Long-term trends in managerial representation. American
Sociological Review, 74, 800 820. doi:10.1177/000312240907400506
*Stoker, J. I. (2007). Sekse en leiderschap: Over het feminiene voordeel en
het masculiene vooroordeel [Gender and leadership: On the feminine
641
advantage and the masculine advantage]. Management & Organisatie,

61, 517.
*Sumer, H. C. (2006). Women in management: Still waiting to be full
members of the club. Sex Roles, 55, 6372. doi:10.1007/s11199-0069059-2
Sutton, A. J. (2009). Publication bias. In H. Cooper, L. V. Hedges, & J. C.
Valentine (Eds.), The handbook of research synthesis and meta-analysis
(2nd ed., pp. 435 452). New York, NY: Russell Sage Foundation.
Sy, T., Shore, L. M., Strauss, J., Shore, T. H., Tram, S., Whiteley, P., &
Ikeda-Muromachi, K. (2010). Leadership perceptions as a function of
race-occupation fit: The case of Asian Americans. Journal of Applied
Psychology, 95, 902919. doi:10.1037/a0019501
*Sylvan, D. L. (1983). The use of sex stereotypes as an information source:
An investigation of sex stereotypes as barriers to women in management.
Dissertation Abstracts International, 44(6-B), 2003.
Tukey, J. W. (1977). Exploratory data analysis. Reading, MA: Addison
Wesley.
United Nations Development Programme. (2006). Human development
report 2006. Retrieved from http://hdr.undp.org/en/media/HDR06complete.pdf
U.S. Courts. (2011). Biographical directory of federal judges. Retrieved
from http://www.uscourts.gov/JudgesAndJudgeships/Biographical
DirectoryOfJudges.aspx
U.S. Department of Labor, Bureau of Labor Statistics. (2007). Historical
CPS employment for 198399 on the 2002 Census industry and occupational classifications. Retrieved from http://www.bls.gov/cps/
constio198399.htm
U.S. Department of Labor, Bureau of Labor Statistics. (2010a). Labor force
statistics from the Current Population Survey. Retrieved from http://
data.bls.gov:8080/PDQ/outside.jsp?survey!ln
U.S. Department of Labor, Bureau of Labor Statistics. (2010b). Women in
the labor force: A databook. Retrieved from http://www.bls.gov/cps/
wlf-databook2010.htm
U.S. Department of Labor, Womens Bureau. (2011). 20 leading occupations of employed women: 2010 annual averages. Retrieved from http://
www.dol.gov/wb/factsheets/20lead2010.htm
*van Engen, M. L. (2006). [Think managerthink male effect]. Unpublished raw data.
*Vinnicombe, S., & Cames, I. (1998). A study of the leadership styles of
female and male managers in 10 different nationality banks in Luxembourg, using the Personal Attributes Questionnaire. International Review
of Women and Leadership, 4, 24 33.
*Vinnicombe, S., & Singh, V. (2002). Sex role stereotyping and requisites
of successful top managers. Women in Management Review, 17, 120
130. doi:10.1108/09649420210425264
von Hippel, W., Sekaquaptewa, D., & Vargas, P. (1995). On the role of
encoding processes in stereotype maintenance. Advances in Experimental Social Psychology, 27, 177254. doi:10.1016/S00652601(08)60406-2
Weber, R., & Crocker, J. (1983). Cognitive processing in the revision of
stereotypic beliefs. Journal of Personality and Social Psychology, 45,
961977. doi:10.1037/0022-3514.45.5.961
Wellington, S., Kropf, M. B., & Gerkovich, P. R. (2003). Whats holding
women back? Harvard Business Review, 81, 18 19.
Widiger, T. A., & Settle, S. A. (1987). Broverman et al. revisited: An
artifactual sex bias. Journal of Personality and Social Psychology, 53,
463 469. doi:10.1037/0022-3514.53.3.463
*Willemsen, T. M. (2002). Gender typing of the successful manager: A
stereotype revisited. Sex Roles, 46, 385391. doi:10.1023/A:
1020409429645
*Williams, W. A. (1989). A study of the degree to which masculinity and
femininity are associated with leadership roles as perceived by educators. Dissertation Abstracts International, 51(6-A), 1871.
Wood, W., & Eagly, A. H. (2010). Gender. In S. T. Fiske, D. T. Gilbert,
642
& G. Lindzey (Eds.), Handbook of social psychology (Vol. 2, 5th ed., pp.
629 667). New York, NY: Wiley.
Wood, W., Lundgren, S., Ouellette, J. A., Busceme, S., & Blackstone, T.
(1994). Minority influence: A meta-analytic review of social influence
processes. Psychological Bulletin, 115, 323345. doi:10.1037/00332909.115.3.323
*Yim, P. C.-Y., & Bond, M. H. (2002). Gender stereotyping of managers
and the self-concept of business students across their undergraduate
education. Women in Management Review, 17, 364 372. doi:10.1108/

09649420210451805
Received December 18, 2009

Revision received February 10, 2011
Accepted March 3, 2011 !

FP Koenig Eagly 2011 PDF

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

FP Koenig Eagly 2011 PDF

Uploaded by

Copyright:

Available Formats

Psychological Bulletin

2011, Vol. 137, No. 4, 616 642

2011 American Psychological Association

Are Leader Stereotypes Masculine?

University of San Diego

Nebraska Wesleyan University

qualities (e.g., assertive, competitive) that they believe are required

The characteristics that people commonly ascribe to women,

The Importance of Stereotypes to Womens Access

This article was published Online First May 30, 2011.

MASCULINITY OF LEADER STEREOTYPES

only of social and organizational psychologists (e.g., Glick &

uative penalty is exacted from women leaders or potential leaders

Three Paradigms for Examining the Masculinity of

KOENIG, EAGLY, MITCHELL, AND RISTIKARI

occupational groups, on a single bipolar masculine versus feminine

Variation in Stereotypes About Men, Women, and

Change in Stereotypes Over Time

MASCULINITY OF LEADER STEREOTYPES

Other Influences on the Masculinity of Leader Roles

percentages of women (vs. men) in leader roles could underlie any

KOENIG, EAGLY, MITCHELL, AND RISTIKARI

92 traits, including many agentic or communal personality traits

The design of the studies in this paradigm appears similar to that of a

MASCULINITY OF LEADER STEREOTYPES

expectation, and perce! (-ption, -tive) or (b) gender terms, such as

Variables Representing Each Study in All Paradigms

For each paradigm, we recorded the exact description of the

Brenner et al. (1989)

Dion & Schuller (1990)

Duehr & Bono (2003)

Duehr & Bono (2006)

Fernandes & Cabral-Cardoso

Fullager et al. (2003)

Heilman et al. (1989)

Karau & Elsaid (2005)

Karau & Hansen (2005)

Boyce & Herd (2003)

Booysen & Nkomo (2006)

Massengill & DiMarco

Neergaard et al. (2007)

Norris & Wylie (1995)

Paradine et al. (1995)

Ryan et al. (in press)

Sauers et al. (2002)

Schein & Davidson

Schein & Mueller

Schein et al. (1989)

Schein et al. (1996)

Kunkel et al. (2003)

MASCULINITY OF LEADER STEREOTYPES

KOENIG, EAGLY, MITCHELL, AND RISTIKARI

Lastly, the source of publication was coded as published or

Computation of Effect Sizes and Data Analysis

Butterfield & Powell (1981)

MASCULINITY OF LEADER STEREOTYPES

Beggs & Doolittle (1993)

MASCULINITY OF LEADER STEREOTYPES

Mahoney & Blake (1987)

MASCULINITY OF LEADER STEREOTYPES

KOENIG, EAGLY, MITCHELL, AND RISTIKARI

variance. In addition, a small studies analysis using cumulative